feat(sidebar): read LibreChat URL from LIBRECHAT_URL env var

Replace hardcoded localhost:3080 chat link with configurable LIBRECHAT_URL environment variable, passed through AuthInfo to the sidebar component. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
feat(librechat): update config URLs for Coolify production deployment
2026-02-24 12:32:57 +01:00 · 2026-02-24 11:41:41 +01:00 · 2026-02-23 22:27:07 +01:00 · 2026-02-23 21:51:46 +01:00 · 2026-02-23 21:50:36 +01:00 · 2026-02-23 21:47:06 +01:00
59 changed files with 504 additions and 4802 deletions
--- a/.env.example
+++ b/.env.example
@@ -34,11 +34,10 @@ MONGODB_DATABASE=certifai
 SEARXNG_URL=http://localhost:8888
 # ---------------------------------------------------------------------------
-# LiteLLM proxy [OPTIONAL - defaults shown]
+# Ollama LLM instance [OPTIONAL - defaults shown]
 # ---------------------------------------------------------------------------
-LITELLM_URL=http://localhost:4000
+OLLAMA_URL=http://localhost:11434
-LITELLM_MODEL=qwen3-32b
+OLLAMA_MODEL=llama3.1:8b
 LITELLM_API_KEY=
 # ---------------------------------------------------------------------------
 # LibreChat (external chat via SSO) [OPTIONAL - default: http://localhost:3080]
@@ -48,7 +47,7 @@ LIBRECHAT_URL=http://localhost:3080
 # ---------------------------------------------------------------------------
 # LLM Providers (comma-separated list) [OPTIONAL]
 # ---------------------------------------------------------------------------
-LLM_PROVIDERS=litellm
+LLM_PROVIDERS=ollama
 # ---------------------------------------------------------------------------
 # SMTP (transactional email) [OPTIONAL]
@@ -67,11 +66,10 @@ STRIPE_WEBHOOK_SECRET=
 STRIPE_PUBLISHABLE_KEY=
 # ---------------------------------------------------------------------------
-# LangChain / LangGraph / LangFlow / Langfuse [OPTIONAL]
+# LangChain / LangGraph / Langfuse [OPTIONAL]
 # ---------------------------------------------------------------------------
 LANGCHAIN_URL=
 LANGGRAPH_URL=
 LANGFLOW_URL=
 LANGFUSE_URL=
 # ---------------------------------------------------------------------------
--- a/.gitea/workflows/ci.yml
+++ b/.gitea/workflows/ci.yml
@@ -120,139 +120,6 @@ jobs:
        run: sccache --show-stats
        if: always()
  # ---------------------------------------------------------------------------
  # Stage 4: E2E tests (only on main, after deploy)
  # ---------------------------------------------------------------------------
  e2e:
    name: E2E Tests
    runs-on: docker
    needs: [deploy]
    if: github.ref == 'refs/heads/main'
    container:
      image: rust:1.89-bookworm
    # MongoDB and SearXNG can start immediately (no repo files needed).
    # Keycloak requires realm-export.json from the repo, so it is started
    # manually after checkout via docker CLI.
    services:
      mongo:
        image: mongo:latest
        env:
          MONGO_INITDB_ROOT_USERNAME: root
          MONGO_INITDB_ROOT_PASSWORD: example
        ports:
          - 27017:27017
      searxng:
        image: searxng/searxng:latest
        env:
          SEARXNG_BASE_URL: http://localhost:8888
        ports:
          - 8888:8080
    env:
      KEYCLOAK_URL: http://localhost:8080
      KEYCLOAK_REALM: certifai
      KEYCLOAK_CLIENT_ID: certifai-dashboard
      MONGODB_URI: mongodb://root:example@mongo:27017
      MONGODB_DATABASE: certifai
      SEARXNG_URL: http://searxng:8080
      LANGGRAPH_URL: ""
      LANGFLOW_URL: ""
      LANGFUSE_URL: ""
    steps:
      - name: Checkout
        run: |
          git init
          git remote add origin "${GITHUB_SERVER_URL}/${GITHUB_REPOSITORY}.git"
          git fetch --depth=1 origin "${GITHUB_SHA}"
          git checkout FETCH_HEAD
      - name: Install system dependencies
        run: |
          apt-get update -qq
          apt-get install -y -qq --no-install-recommends \
            unzip curl docker.io \
            libglib2.0-0 libnss3 libnspr4 libdbus-1-3 libatk1.0-0 \
            libatk-bridge2.0-0 libcups2 libdrm2 libxkbcommon0 libxcomposite1 \
            libxdamage1 libxfixes3 libxrandr2 libgbm1 libpango-1.0-0 \
            libcairo2 libasound2 libatspi2.0-0 libxshmfence1
      - name: Start Keycloak
        run: |
          docker run -d --name ci-keycloak --network host \
            -e KC_BOOTSTRAP_ADMIN_USERNAME=admin \
            -e KC_BOOTSTRAP_ADMIN_PASSWORD=admin \
            -e KC_DB=dev-mem \
            -e KC_HEALTH_ENABLED=true \
            -v "$PWD/keycloak/realm-export.json:/opt/keycloak/data/import/realm-export.json:ro" \
            -v "$PWD/keycloak/themes/certifai:/opt/keycloak/themes/certifai:ro" \
            quay.io/keycloak/keycloak:26.0 start-dev --import-realm
          echo "Waiting for Keycloak..."
          for i in $(seq 1 60); do
            if curl -sf http://localhost:8080/realms/certifai > /dev/null 2>&1; then
              echo "Keycloak is ready"
              break
            fi
            if [ "$i" -eq 60 ]; then
              echo "Keycloak failed to start within 60s"
              docker logs ci-keycloak
              exit 1
            fi
            sleep 2
          done
      - name: Install sccache
        run: |
          curl -fsSL https://github.com/mozilla/sccache/releases/download/v0.9.1/sccache-v0.9.1-x86_64-unknown-linux-musl.tar.gz \
            | tar xz --strip-components=1 -C /usr/local/bin/ sccache-v0.9.1-x86_64-unknown-linux-musl/sccache
          chmod +x /usr/local/bin/sccache
      - name: Install dioxus-cli
        run: cargo install dioxus-cli --locked
      - name: Install bun
        run: |
          curl -fsSL https://bun.sh/install | bash
          echo "$HOME/.bun/bin" >> "$GITHUB_PATH"
      - name: Install Playwright
        run: |
          export PATH="$HOME/.bun/bin:$PATH"
          bun install
          bunx playwright install chromium
      - name: Build app
        run: dx build --release
      - name: Start app and run E2E tests
        run: |
          export PATH="$HOME/.bun/bin:$PATH"
          # Start the app in the background
          dx serve --release --port 8000 &
          APP_PID=$!
          # Wait for the app to be ready
          echo "Waiting for app to start..."
          for i in $(seq 1 60); do
            if curl -sf http://localhost:8000 > /dev/null 2>&1; then
              echo "App is ready"
              break
            fi
            if [ "$i" -eq 60 ]; then
              echo "App failed to start within 60s"
              exit 1
            fi
            sleep 1
          done
          BASE_URL=http://localhost:8000 bunx playwright test --reporter=list
          kill "$APP_PID" 2>/dev/null || true
      - name: Upload test report
        if: always()
        uses: actions/upload-artifact@v4
        with:
          name: playwright-report
          path: playwright-report/
          retention-days: 7
      - name: Cleanup Keycloak
        if: always()
        run: docker rm -f ci-keycloak 2>/dev/null || true
      - name: Show sccache stats
        run: sccache --show-stats
        if: always()
  # ---------------------------------------------------------------------------
  # Stage 3: Deploy (only after tests pass, only on main)
  # ---------------------------------------------------------------------------
--- a/.gitignore
+++ b/.gitignore
@@ -22,8 +22,3 @@ keycloak/*
 node_modules/
 searxng/
 # Playwright
 e2e/.auth/
 playwright-report/
 test-results/
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -773,11 +773,9 @@ dependencies = [
 "dioxus-sdk",
 "dotenvy",
 "futures",
 "js-sys",
 "maud",
 "mongodb",
 "petname",
 "pretty_assertions",
 "pulldown-cmark",
 "rand 0.10.0",
 "reqwest 0.13.2",
@@ -785,7 +783,6 @@ dependencies = [
 "secrecy",
 "serde",
 "serde_json",
 "serial_test",
 "sha2",
 "thiserror 2.0.18",
 "time",
@@ -885,12 +882,6 @@ dependencies = [
 "unicode-xid",
 ]
 [[package]]
 name = "diff"
 version = "0.1.13"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "56254986775e3233ffa9c4d7d3faaf6d36a2c09d30b20687e9f88bc8bafc16c8"
 [[package]]
 name = "digest"
 version = "0.10.7"
@@ -3255,16 +3246,6 @@ version = "0.1.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "925383efa346730478fb4838dbe9137d2a47675ad789c546d150a6e1dd4ab31c"
 [[package]]
 name = "pretty_assertions"
 version = "1.4.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3ae130e2f271fbc2ac3a40fb1d07180839cdbbe443c7a27e1e3c13c5cac0116d"
 dependencies = [
 "diff",
 "yansi",
 ]
 [[package]]
 name = "prettyplease"
 version = "0.2.37"
@@ -3842,15 +3823,6 @@ dependencies = [
 "winapi-util",
 ]
 [[package]]
 name = "scc"
 version = "2.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "46e6f046b7fef48e2660c57ed794263155d713de679057f2d0c169bfc6e756cc"
 dependencies = [
 "sdd",
 ]
 [[package]]
 name = "schannel"
 version = "0.1.28"
@@ -3890,12 +3862,6 @@ dependencies = [
 "untrusted",
 ]
 [[package]]
 name = "sdd"
 version = "3.0.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "490dcfcbfef26be6800d11870ff2df8774fa6e86d047e3e8c8a76b25655e41ca"
 [[package]]
 name = "secrecy"
 version = "0.10.3"
@@ -4116,32 +4082,6 @@ dependencies = [
 "syn 2.0.116",
 ]
 [[package]]
 name = "serial_test"
 version = "3.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "911bd979bf1070a3f3aa7b691a3b3e9968f339ceeec89e08c280a8a22207a32f"
 dependencies = [
 "futures-executor",
 "futures-util",
 "log",
 "once_cell",
 "parking_lot",
 "scc",
 "serial_test_derive",
 ]
 [[package]]
 name = "serial_test_derive"
 version = "3.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "0a7d91949b85b0d2fb687445e448b40d322b6b3e4af6b44a29b21d9a5f33e6d9"
 dependencies = [
 "proc-macro2",
 "quote",
 "syn 2.0.116",
 ]
 [[package]]
 name = "servo_arc"
 version = "0.4.3"
@@ -5743,12 +5683,6 @@ version = "0.8.15"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "fdd20c5420375476fbd4394763288da7eb0cc0b8c11deed431a91562af7335d3"
 [[package]]
 name = "yansi"
 version = "1.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "cfe53a6657fd280eaa890a3bc59152892ffa3e30101319d168b781ed6529b049"
 [[package]]
 name = "yazi"
 version = "0.1.6"
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -61,7 +61,6 @@ secrecy = { version = "0.10", default-features = false, optional = true }
 serde_json = { version = "1.0.133", default-features = false }
 maud = { version = "0.27", default-features = false }
 url = { version = "2.5.4", default-features = false, optional = true }
 js-sys = { version = "0.3", optional = true }
 wasm-bindgen = { version = "0.2", optional = true }
 web-sys = { version = "0.3", optional = true, features = [
    "Clipboard",
@@ -92,7 +91,7 @@ bytes = { version = "1", optional = true }
 [features]
 # default = ["web"]
-web = ["dioxus/web", "dep:reqwest", "dep:web-sys", "dep:wasm-bindgen", "dep:js-sys"]
+web = ["dioxus/web", "dep:reqwest", "dep:web-sys", "dep:wasm-bindgen"]
 server = [
    "dioxus/server",
    "dep:axum",
@@ -113,10 +112,6 @@ server = [
    "dep:bytes",
 ]
 [dev-dependencies]
 pretty_assertions = "1.4"
 serial_test = "3.2"
 [[bin]]
 name = "dashboard"
 path = "bin/main.rs"
--- a/assets/i18n/de.json
+++ b/assets/i18n/de.json
@@ -58,15 +58,15 @@
    "title": "Dashboard",
    "subtitle": "KI-Nachrichten und Neuigkeiten",
    "topic_placeholder": "Themenname...",
-    "litellm_settings": "LiteLLM-Einstellungen",
+    "ollama_settings": "Ollama-Einstellungen",
-    "settings_hint": "Leer lassen, um LITELLM_URL / LITELLM_MODEL aus .env zu verwenden",
+    "settings_hint": "Leer lassen, um OLLAMA_URL / OLLAMA_MODEL aus .env zu verwenden",
-    "litellm_url": "LiteLLM-URL",
+    "ollama_url": "Ollama-URL",
-    "litellm_url_placeholder": "Verwendet LITELLM_URL aus .env",
+    "ollama_url_placeholder": "Verwendet OLLAMA_URL aus .env",
    "model": "Modell",
-    "model_placeholder": "Verwendet LITELLM_MODEL aus .env",
+    "model_placeholder": "Verwendet OLLAMA_MODEL aus .env",
    "searching": "Suche laeuft...",
    "search_failed": "Suche fehlgeschlagen: {e}",
-    "litellm_status": "LiteLLM-Status",
+    "ollama_status": "Ollama-Status",
    "trending": "Im Trend",
    "recent_searches": "Letzte Suchen"
  },
@@ -96,38 +96,7 @@
    "total_requests": "Anfragen gesamt",
    "avg_latency": "Durchschn. Latenz",
    "tokens_used": "Verbrauchte Token",
-    "error_rate": "Fehlerrate",
+    "error_rate": "Fehlerrate"
    "not_configured": "Nicht konfiguriert",
    "open_new_tab": "In neuem Tab oeffnen",
    "agents_status_connected": "Verbunden",
    "agents_status_not_connected": "Nicht verbunden",
    "agents_config_hint": "Setzen Sie LANGGRAPH_URL in .env, um eine Verbindung herzustellen",
    "agents_quick_start": "Schnellstart",
    "agents_docs": "Dokumentation",
    "agents_docs_desc": "Offizielle LangGraph-Dokumentation und API-Anleitungen.",
    "agents_getting_started": "Erste Schritte",
    "agents_getting_started_desc": "Schritt-fuer-Schritt-Anleitung zum Erstellen Ihres ersten Agenten.",
    "agents_github": "GitHub",
    "agents_github_desc": "Quellcode, Issues und Community-Beitraege.",
    "agents_examples": "Beispiele",
    "agents_examples_desc": "Einsatzbereite Vorlagen und Beispielprojekte fuer Agenten.",
    "agents_api_ref": "API-Referenz",
    "agents_api_ref_desc": "Lokale Swagger-Dokumentation fuer Ihre LangGraph-Instanz.",
    "agents_running_title": "Laufende Agenten",
    "agents_none": "Keine Agenten registriert. Stellen Sie einen Assistenten in LangGraph bereit, um ihn hier zu sehen.",
    "agents_col_name": "Name",
    "agents_col_id": "ID",
    "agents_col_description": "Beschreibung",
    "agents_col_status": "Status",
    "analytics_status_connected": "Verbunden",
    "analytics_status_not_connected": "Nicht verbunden",
    "analytics_config_hint": "Setzen Sie LANGFUSE_URL in .env, um eine Verbindung herzustellen",
    "analytics_sso_hint": "Langfuse nutzt Keycloak-SSO. Sie werden automatisch mit Ihrem CERTifAI-Konto angemeldet.",
    "analytics_quick_actions": "Schnellaktionen",
    "analytics_traces": "Traces",
    "analytics_traces_desc": "Alle LLM-Aufrufe, Latenzen und Token-Verbrauch anzeigen und filtern.",
    "analytics_dashboard": "Dashboard",
    "analytics_dashboard_desc": "Ueberblick ueber Kosten, Qualitaetsmetriken und Nutzungstrends."
  },
  "org": {
    "title": "Organisation",
@@ -144,16 +113,6 @@
    "email_address": "E-Mail-Adresse",
    "email_placeholder": "kollege@firma.de",
    "send_invite": "Einladung senden",
    "total_spend": "Gesamtausgaben",
    "total_tokens": "Tokens gesamt",
    "model_usage": "Nutzung nach Modell",
    "model": "Modell",
    "tokens": "Tokens",
    "spend": "Ausgaben",
    "usage_unavailable": "Nutzungsdaten nicht verfuegbar",
    "loading_usage": "Nutzungsdaten werden geladen...",
    "prompt_tokens": "Prompt-Tokens",
    "completion_tokens": "Antwort-Tokens",
    "pricing_title": "Preise",
    "pricing_subtitle": "Waehlen Sie den passenden Plan fuer Ihre Organisation"
  },
--- a/assets/i18n/en.json
+++ b/assets/i18n/en.json
@@ -58,15 +58,15 @@
    "title": "Dashboard",
    "subtitle": "AI news and updates",
    "topic_placeholder": "Topic name...",
-    "litellm_settings": "LiteLLM Settings",
+    "ollama_settings": "Ollama Settings",
-    "settings_hint": "Leave empty to use LITELLM_URL / LITELLM_MODEL from .env",
+    "settings_hint": "Leave empty to use OLLAMA_URL / OLLAMA_MODEL from .env",
-    "litellm_url": "LiteLLM URL",
+    "ollama_url": "Ollama URL",
-    "litellm_url_placeholder": "Uses LITELLM_URL from .env",
+    "ollama_url_placeholder": "Uses OLLAMA_URL from .env",
    "model": "Model",
-    "model_placeholder": "Uses LITELLM_MODEL from .env",
+    "model_placeholder": "Uses OLLAMA_MODEL from .env",
    "searching": "Searching...",
    "search_failed": "Search failed: {e}",
-    "litellm_status": "LiteLLM Status",
+    "ollama_status": "Ollama Status",
    "trending": "Trending",
    "recent_searches": "Recent Searches"
  },
@@ -96,38 +96,7 @@
    "total_requests": "Total Requests",
    "avg_latency": "Avg Latency",
    "tokens_used": "Tokens Used",
-    "error_rate": "Error Rate",
+    "error_rate": "Error Rate"
    "not_configured": "Not Configured",
    "open_new_tab": "Open in New Tab",
    "agents_status_connected": "Connected",
    "agents_status_not_connected": "Not Connected",
    "agents_config_hint": "Set LANGGRAPH_URL in .env to connect",
    "agents_quick_start": "Quick Start",
    "agents_docs": "Documentation",
    "agents_docs_desc": "Official LangGraph documentation and API guides.",
    "agents_getting_started": "Getting Started",
    "agents_getting_started_desc": "Step-by-step tutorial to build your first agent.",
    "agents_github": "GitHub",
    "agents_github_desc": "Source code, issues, and community contributions.",
    "agents_examples": "Examples",
    "agents_examples_desc": "Ready-to-use templates and example agent projects.",
    "agents_api_ref": "API Reference",
    "agents_api_ref_desc": "Local Swagger docs for your LangGraph instance.",
    "agents_running_title": "Running Agents",
    "agents_none": "No agents registered. Deploy an assistant to LangGraph to see it here.",
    "agents_col_name": "Name",
    "agents_col_id": "ID",
    "agents_col_description": "Description",
    "agents_col_status": "Status",
    "analytics_status_connected": "Connected",
    "analytics_status_not_connected": "Not Connected",
    "analytics_config_hint": "Set LANGFUSE_URL in .env to connect",
    "analytics_sso_hint": "Langfuse uses Keycloak SSO. You will be signed in automatically with your CERTifAI account.",
    "analytics_quick_actions": "Quick Actions",
    "analytics_traces": "Traces",
    "analytics_traces_desc": "View and filter all LLM call traces, latencies, and token usage.",
    "analytics_dashboard": "Dashboard",
    "analytics_dashboard_desc": "Overview of costs, quality metrics, and usage trends."
  },
  "org": {
    "title": "Organization",
@@ -144,16 +113,6 @@
    "email_address": "Email Address",
    "email_placeholder": "colleague@company.com",
    "send_invite": "Send Invite",
    "total_spend": "Total Spend",
    "total_tokens": "Total Tokens",
    "model_usage": "Usage by Model",
    "model": "Model",
    "tokens": "Tokens",
    "spend": "Spend",
    "usage_unavailable": "Usage data unavailable",
    "loading_usage": "Loading usage data...",
    "prompt_tokens": "Prompt Tokens",
    "completion_tokens": "Completion Tokens",
    "pricing_title": "Pricing",
    "pricing_subtitle": "Choose the plan that fits your organization"
  },
--- a/assets/i18n/es.json
+++ b/assets/i18n/es.json
@@ -58,15 +58,15 @@
    "title": "Panel de control",
    "subtitle": "Noticias y actualizaciones de IA",
    "topic_placeholder": "Nombre del tema...",
-    "litellm_settings": "Configuracion de LiteLLM",
+    "ollama_settings": "Configuracion de Ollama",
-    "settings_hint": "Dejar vacio para usar LITELLM_URL / LITELLM_MODEL del archivo .env",
+    "settings_hint": "Dejar vacio para usar OLLAMA_URL / OLLAMA_MODEL del archivo .env",
-    "litellm_url": "URL de LiteLLM",
+    "ollama_url": "URL de Ollama",
-    "litellm_url_placeholder": "Usa LITELLM_URL del archivo .env",
+    "ollama_url_placeholder": "Usa OLLAMA_URL del archivo .env",
    "model": "Modelo",
-    "model_placeholder": "Usa LITELLM_MODEL del archivo .env",
+    "model_placeholder": "Usa OLLAMA_MODEL del archivo .env",
    "searching": "Buscando...",
    "search_failed": "La busqueda fallo: {e}",
-    "litellm_status": "Estado de LiteLLM",
+    "ollama_status": "Estado de Ollama",
    "trending": "Tendencias",
    "recent_searches": "Busquedas recientes"
  },
@@ -96,38 +96,7 @@
    "total_requests": "Total de solicitudes",
    "avg_latency": "Latencia promedio",
    "tokens_used": "Tokens utilizados",
-    "error_rate": "Tasa de errores",
+    "error_rate": "Tasa de errores"
    "not_configured": "No configurado",
    "open_new_tab": "Abrir en nueva pestana",
    "agents_status_connected": "Conectado",
    "agents_status_not_connected": "No conectado",
    "agents_config_hint": "Configure LANGGRAPH_URL en .env para conectar",
    "agents_quick_start": "Inicio rapido",
    "agents_docs": "Documentacion",
    "agents_docs_desc": "Documentacion oficial de LangGraph y guias de API.",
    "agents_getting_started": "Primeros pasos",
    "agents_getting_started_desc": "Tutorial paso a paso para crear su primer agente.",
    "agents_github": "GitHub",
    "agents_github_desc": "Codigo fuente, issues y contribuciones de la comunidad.",
    "agents_examples": "Ejemplos",
    "agents_examples_desc": "Plantillas y proyectos de agentes listos para usar.",
    "agents_api_ref": "Referencia API",
    "agents_api_ref_desc": "Documentacion Swagger local para su instancia de LangGraph.",
    "agents_running_title": "Agentes en ejecucion",
    "agents_none": "No hay agentes registrados. Despliegue un asistente en LangGraph para verlo aqui.",
    "agents_col_name": "Nombre",
    "agents_col_id": "ID",
    "agents_col_description": "Descripcion",
    "agents_col_status": "Estado",
    "analytics_status_connected": "Conectado",
    "analytics_status_not_connected": "No conectado",
    "analytics_config_hint": "Configure LANGFUSE_URL en .env para conectar",
    "analytics_sso_hint": "Langfuse utiliza SSO de Keycloak. Iniciara sesion automaticamente con su cuenta CERTifAI.",
    "analytics_quick_actions": "Acciones rapidas",
    "analytics_traces": "Trazas",
    "analytics_traces_desc": "Ver y filtrar todas las llamadas LLM, latencias y uso de tokens.",
    "analytics_dashboard": "Panel de control",
    "analytics_dashboard_desc": "Resumen de costos, metricas de calidad y tendencias de uso."
  },
  "org": {
    "title": "Organizacion",
@@ -144,16 +113,6 @@
    "email_address": "Direccion de correo electronico",
    "email_placeholder": "colega@empresa.com",
    "send_invite": "Enviar invitacion",
    "total_spend": "Gasto total",
    "total_tokens": "Tokens totales",
    "model_usage": "Uso por modelo",
    "model": "Modelo",
    "tokens": "Tokens",
    "spend": "Gasto",
    "usage_unavailable": "Datos de uso no disponibles",
    "loading_usage": "Cargando datos de uso...",
    "prompt_tokens": "Tokens de entrada",
    "completion_tokens": "Tokens de respuesta",
    "pricing_title": "Precios",
    "pricing_subtitle": "Elija el plan que se adapte a su organizacion"
  },
--- a/assets/i18n/fr.json
+++ b/assets/i18n/fr.json
@@ -58,15 +58,15 @@
    "title": "Tableau de bord",
    "subtitle": "Actualites et mises a jour IA",
    "topic_placeholder": "Nom du sujet...",
-    "litellm_settings": "Parametres LiteLLM",
+    "ollama_settings": "Parametres Ollama",
-    "settings_hint": "Laissez vide pour utiliser LITELLM_URL / LITELLM_MODEL du fichier .env",
+    "settings_hint": "Laissez vide pour utiliser OLLAMA_URL / OLLAMA_MODEL du fichier .env",
-    "litellm_url": "URL LiteLLM",
+    "ollama_url": "URL Ollama",
-    "litellm_url_placeholder": "Utilise LITELLM_URL du fichier .env",
+    "ollama_url_placeholder": "Utilise OLLAMA_URL du fichier .env",
    "model": "Modele",
-    "model_placeholder": "Utilise LITELLM_MODEL du fichier .env",
+    "model_placeholder": "Utilise OLLAMA_MODEL du fichier .env",
    "searching": "Recherche en cours...",
    "search_failed": "Echec de la recherche : {e}",
-    "litellm_status": "Statut LiteLLM",
+    "ollama_status": "Statut Ollama",
    "trending": "Tendances",
    "recent_searches": "Recherches recentes"
  },
@@ -96,38 +96,7 @@
    "total_requests": "Requetes totales",
    "avg_latency": "Latence moyenne",
    "tokens_used": "Tokens utilises",
-    "error_rate": "Taux d'erreur",
+    "error_rate": "Taux d'erreur"
    "not_configured": "Non configure",
    "open_new_tab": "Ouvrir dans un nouvel onglet",
    "agents_status_connected": "Connecte",
    "agents_status_not_connected": "Non connecte",
    "agents_config_hint": "Definissez LANGGRAPH_URL dans .env pour vous connecter",
    "agents_quick_start": "Demarrage rapide",
    "agents_docs": "Documentation",
    "agents_docs_desc": "Documentation officielle de LangGraph et guides API.",
    "agents_getting_started": "Premiers pas",
    "agents_getting_started_desc": "Tutoriel etape par etape pour creer votre premier agent.",
    "agents_github": "GitHub",
    "agents_github_desc": "Code source, issues et contributions de la communaute.",
    "agents_examples": "Exemples",
    "agents_examples_desc": "Modeles et projets d'agents prets a l'emploi.",
    "agents_api_ref": "Reference API",
    "agents_api_ref_desc": "Documentation Swagger locale pour votre instance LangGraph.",
    "agents_running_title": "Agents en cours",
    "agents_none": "Aucun agent enregistre. Deployez un assistant dans LangGraph pour le voir ici.",
    "agents_col_name": "Nom",
    "agents_col_id": "ID",
    "agents_col_description": "Description",
    "agents_col_status": "Statut",
    "analytics_status_connected": "Connecte",
    "analytics_status_not_connected": "Non connecte",
    "analytics_config_hint": "Definissez LANGFUSE_URL dans .env pour vous connecter",
    "analytics_sso_hint": "Langfuse utilise le SSO Keycloak. Vous serez connecte automatiquement avec votre compte CERTifAI.",
    "analytics_quick_actions": "Actions rapides",
    "analytics_traces": "Traces",
    "analytics_traces_desc": "Afficher et filtrer tous les appels LLM, latences et consommation de tokens.",
    "analytics_dashboard": "Tableau de bord",
    "analytics_dashboard_desc": "Apercu des couts, metriques de qualite et tendances d'utilisation."
  },
  "org": {
    "title": "Organisation",
@@ -144,16 +113,6 @@
    "email_address": "Adresse e-mail",
    "email_placeholder": "collegue@entreprise.com",
    "send_invite": "Envoyer l'invitation",
    "total_spend": "Depenses totales",
    "total_tokens": "Tokens totaux",
    "model_usage": "Utilisation par modele",
    "model": "Modele",
    "tokens": "Tokens",
    "spend": "Depenses",
    "usage_unavailable": "Donnees d'utilisation indisponibles",
    "loading_usage": "Chargement des donnees d'utilisation...",
    "prompt_tokens": "Tokens d'entree",
    "completion_tokens": "Tokens de reponse",
    "pricing_title": "Tarifs",
    "pricing_subtitle": "Choisissez le plan adapte a votre organisation"
  },
--- a/assets/i18n/pt.json
+++ b/assets/i18n/pt.json
@@ -58,15 +58,15 @@
    "title": "Painel",
    "subtitle": "Noticias e atualizacoes de IA",
    "topic_placeholder": "Nome do topico...",
-    "litellm_settings": "Definicoes do LiteLLM",
+    "ollama_settings": "Definicoes do Ollama",
-    "settings_hint": "Deixe vazio para usar LITELLM_URL / LITELLM_MODEL do .env",
+    "settings_hint": "Deixe vazio para usar OLLAMA_URL / OLLAMA_MODEL do .env",
-    "litellm_url": "URL do LiteLLM",
+    "ollama_url": "URL do Ollama",
-    "litellm_url_placeholder": "Utiliza LITELLM_URL do .env",
+    "ollama_url_placeholder": "Utiliza OLLAMA_URL do .env",
    "model": "Modelo",
-    "model_placeholder": "Utiliza LITELLM_MODEL do .env",
+    "model_placeholder": "Utiliza OLLAMA_MODEL do .env",
    "searching": "A pesquisar...",
    "search_failed": "A pesquisa falhou: {e}",
-    "litellm_status": "Estado do LiteLLM",
+    "ollama_status": "Estado do Ollama",
    "trending": "Em destaque",
    "recent_searches": "Pesquisas recentes"
  },
@@ -96,38 +96,7 @@
    "total_requests": "Total de Pedidos",
    "avg_latency": "Latencia Media",
    "tokens_used": "Tokens Utilizados",
-    "error_rate": "Taxa de Erros",
+    "error_rate": "Taxa de Erros"
    "not_configured": "Nao configurado",
    "open_new_tab": "Abrir em novo separador",
    "agents_status_connected": "Conectado",
    "agents_status_not_connected": "Nao conectado",
    "agents_config_hint": "Defina LANGGRAPH_URL no .env para conectar",
    "agents_quick_start": "Inicio rapido",
    "agents_docs": "Documentacao",
    "agents_docs_desc": "Documentacao oficial do LangGraph e guias de API.",
    "agents_getting_started": "Primeiros passos",
    "agents_getting_started_desc": "Tutorial passo a passo para criar o seu primeiro agente.",
    "agents_github": "GitHub",
    "agents_github_desc": "Codigo fonte, issues e contribuicoes da comunidade.",
    "agents_examples": "Exemplos",
    "agents_examples_desc": "Modelos e projetos de agentes prontos a usar.",
    "agents_api_ref": "Referencia API",
    "agents_api_ref_desc": "Documentacao Swagger local para a sua instancia LangGraph.",
    "agents_running_title": "Agentes em execucao",
    "agents_none": "Nenhum agente registado. Implemente um assistente no LangGraph para o ver aqui.",
    "agents_col_name": "Nome",
    "agents_col_id": "ID",
    "agents_col_description": "Descricao",
    "agents_col_status": "Estado",
    "analytics_status_connected": "Conectado",
    "analytics_status_not_connected": "Nao conectado",
    "analytics_config_hint": "Defina LANGFUSE_URL no .env para conectar",
    "analytics_sso_hint": "O Langfuse utiliza SSO do Keycloak. Sera autenticado automaticamente com a sua conta CERTifAI.",
    "analytics_quick_actions": "Acoes rapidas",
    "analytics_traces": "Traces",
    "analytics_traces_desc": "Ver e filtrar todas as chamadas LLM, latencias e uso de tokens.",
    "analytics_dashboard": "Painel",
    "analytics_dashboard_desc": "Resumo de custos, metricas de qualidade e tendencias de uso."
  },
  "org": {
    "title": "Organizacao",
@@ -144,16 +113,6 @@
    "email_address": "Endereco de Email",
    "email_placeholder": "colleague@company.com",
    "send_invite": "Enviar Convite",
    "total_spend": "Gasto total",
    "total_tokens": "Tokens totais",
    "model_usage": "Uso por modelo",
    "model": "Modelo",
    "tokens": "Tokens",
    "spend": "Gasto",
    "usage_unavailable": "Dados de uso indisponiveis",
    "loading_usage": "Carregando dados de uso...",
    "prompt_tokens": "Tokens de entrada",
    "completion_tokens": "Tokens de resposta",
    "pricing_title": "Precos",
    "pricing_subtitle": "Escolha o plano adequado a sua organizacao"
  },
--- a/assets/main.css
+++ b/assets/main.css
@@ -2591,58 +2591,6 @@ h6 {
    border-radius: 20px;
 }
 /* ===== Tool Embed (iframe integration) ===== */
 .tool-embed {
    display: flex;
    flex-direction: column;
    flex: 1;
    height: calc(100vh - 60px);
    min-height: 400px;
 }
 .tool-embed-toolbar {
    display: flex;
    align-items: center;
    justify-content: space-between;
    padding: 12px 20px;
    background-color: var(--bg-card);
    border-bottom: 1px solid var(--border-primary);
 }
 .tool-embed-title {
    font-family: 'Space Grotesk', sans-serif;
    font-size: 16px;
    font-weight: 600;
    color: var(--text-heading);
 }
 .tool-embed-popout-btn {
    display: inline-flex;
    align-items: center;
    gap: 6px;
    padding: 6px 14px;
    font-size: 13px;
    font-weight: 500;
    color: var(--accent);
    background-color: transparent;
    border: 1px solid var(--accent);
    border-radius: 6px;
    text-decoration: none;
    cursor: pointer;
    transition: background-color 0.15s, color 0.15s;
 }
 .tool-embed-popout-btn:hover {
    background-color: var(--accent);
    color: var(--bg-body);
 }
 .tool-embed-iframe {
    flex: 1;
    width: 100%;
    border: none;
 }
 /* ===== Analytics Stats Bar ===== */
 .analytics-stats-bar {
    display: flex;
@@ -3375,341 +3323,3 @@ h6 {
        padding: 20px 16px;
    }
 }
 /* ===== Agents Page ===== */
 .agents-page {
    display: flex;
    flex-direction: column;
    padding: 32px;
    gap: 32px;
 }
 .agents-hero {
    max-width: 720px;
    display: flex;
    flex-direction: column;
    gap: 12px;
 }
 .agents-hero-row {
    display: flex;
    flex-direction: row;
    align-items: center;
    gap: 16px;
 }
 .agents-hero-icon {
    width: 48px;
    height: 48px;
    min-width: 48px;
    background: linear-gradient(135deg, var(--accent), var(--accent-secondary));
    color: var(--avatar-text);
    border-radius: 12px;
    font-size: 24px;
    display: flex;
    align-items: center;
    justify-content: center;
    font-weight: 700;
 }
 .agents-hero-title {
    font-family: 'Space Grotesk', sans-serif;
    font-size: 28px;
    font-weight: 700;
    color: var(--text-heading);
    margin: 0;
 }
 .agents-hero-desc {
    font-size: 15px;
    color: var(--text-muted);
    line-height: 1.6;
    max-width: 600px;
    margin: 0;
 }
 .agents-status {
    display: inline-flex;
    align-items: center;
    gap: 8px;
    font-size: 13px;
 }
 .agents-status-dot {
    width: 8px;
    height: 8px;
    border-radius: 50%;
    display: inline-block;
    flex-shrink: 0;
 }
 .agents-status-dot--on {
    background-color: #22c55e;
 }
 .agents-status-dot--off {
    background-color: var(--text-faint);
 }
 .agents-status-url {
    font-family: 'JetBrains Mono', 'Fira Code', monospace;
    color: var(--accent);
    font-size: 13px;
 }
 .agents-status-hint {
    font-size: 13px;
    color: var(--text-faint);
    font-style: italic;
 }
 .agents-section-title {
    font-size: 18px;
    font-weight: 600;
    color: var(--text-heading);
    margin: 0 0 12px 0;
 }
 .agents-grid {
    display: grid;
    grid-template-columns: repeat(3, 1fr);
    gap: 16px;
 }
 .agents-card {
    display: block;
    text-decoration: none;
    background-color: var(--bg-card);
    border: 1px solid var(--border-primary);
    border-radius: 12px;
    padding: 24px;
    transition: border-color 0.2s, transform 0.2s;
    cursor: pointer;
 }
 .agents-card:hover {
    border-color: var(--accent);
    transform: translateY(-2px);
 }
 .agents-card-icon {
    width: 36px;
    height: 36px;
    min-width: 36px;
    background: linear-gradient(135deg, var(--accent), var(--accent-secondary));
    color: var(--avatar-text);
    border-radius: 8px;
    font-size: 18px;
    display: flex;
    align-items: center;
    justify-content: center;
    font-weight: 700;
 }
 .agents-card-title {
    font-size: 16px;
    font-weight: 600;
    color: var(--text-heading);
    margin: 12px 0 4px;
 }
 .agents-card-desc {
    font-size: 13px;
    color: var(--text-muted);
    line-height: 1.5;
 }
 .agents-card--disabled {
    opacity: 0.4;
    pointer-events: none;
    cursor: default;
 }
 /* -- Agents table -- */
 .agents-table-section {
    max-width: 960px;
 }
 .agents-table-wrap {
    overflow-x: auto;
 }
 .agents-table {
    width: 100%;
    border-collapse: collapse;
    font-size: 14px;
 }
 .agents-table thead th {
    text-align: left;
    font-size: 12px;
    font-weight: 600;
    color: var(--text-faint);
    text-transform: uppercase;
    letter-spacing: 0.04em;
    padding: 8px 12px;
    border-bottom: 1px solid var(--border-secondary);
 }
 .agents-table tbody td {
    padding: 10px 12px;
    border-bottom: 1px solid var(--border-primary);
    color: var(--text-primary);
    vertical-align: middle;
 }
 .agents-table tbody tr:hover {
    background-color: var(--bg-surface);
 }
 .agents-cell-name {
    font-weight: 600;
    color: var(--text-heading);
    white-space: nowrap;
 }
 .agents-cell-id {
    font-family: 'JetBrains Mono', 'Fira Code', monospace;
    font-size: 12px;
    color: var(--text-muted);
 }
 .agents-cell-desc {
    max-width: 300px;
    overflow: hidden;
    text-overflow: ellipsis;
    white-space: nowrap;
    color: var(--text-muted);
 }
 .agents-cell-none {
    color: var(--text-faint);
 }
 .agents-badge {
    display: inline-block;
    font-size: 12px;
    font-weight: 600;
    padding: 2px 10px;
    border-radius: 9999px;
 }
 .agents-badge--active {
    background-color: rgba(34, 197, 94, 0.15);
    color: #22c55e;
 }
 .agents-table-loading,
 .agents-table-empty {
    font-size: 14px;
    color: var(--text-faint);
    font-style: italic;
    padding: 16px 0;
 }
@media (max-width: 768px) {
    .agents-grid {
        grid-template-columns: repeat(2, 1fr);
    }
 }
@media (max-width: 480px) {
    .agents-page,
    .analytics-page {
        padding: 20px 16px;
    }
    .agents-grid {
        grid-template-columns: 1fr;
    }
 }
 /* ===== Analytics Page ===== */
 .analytics-page {
    display: flex;
    flex-direction: column;
    padding: 32px;
    gap: 32px;
 }
 .analytics-hero {
    max-width: 720px;
    display: flex;
    flex-direction: column;
    gap: 12px;
 }
 .analytics-hero-row {
    display: flex;
    flex-direction: row;
    align-items: center;
    gap: 16px;
 }
 .analytics-hero-icon {
    width: 48px;
    height: 48px;
    min-width: 48px;
    background: linear-gradient(135deg, var(--accent), var(--accent-secondary));
    color: var(--avatar-text);
    border-radius: 12px;
    font-size: 24px;
    display: flex;
    align-items: center;
    justify-content: center;
 }
 .analytics-hero-title {
    font-family: 'Space Grotesk', sans-serif;
    font-size: 28px;
    font-weight: 700;
    color: var(--text-heading);
    margin: 0;
 }
 .analytics-hero-desc {
    font-size: 15px;
    color: var(--text-muted);
    line-height: 1.6;
    max-width: 600px;
    margin: 0;
 }
 .analytics-sso-hint {
    font-size: 13px;
    color: var(--text-muted);
    font-style: italic;
    margin: 0;
 }
 .analytics-launch-btn {
    display: inline-flex;
    align-items: center;
    gap: 8px;
    padding: 10px 20px;
    background: linear-gradient(135deg, var(--accent), var(--accent-secondary));
    color: var(--avatar-text);
    font-size: 14px;
    font-weight: 600;
    border-radius: 8px;
    text-decoration: none;
    transition: opacity 0.2s, transform 0.2s;
    width: fit-content;
 }
 .analytics-launch-btn:hover {
    opacity: 0.9;
    transform: translateY(-1px);
 }
 .analytics-stats-bar {
    display: flex;
    gap: 16px;
    flex-wrap: wrap;
 }
@media (max-width: 768px) {
    .analytics-stats-bar {
        flex-direction: column;
    }
 }
--- a/assets/tailwind.css
+++ b/assets/tailwind.css
@@ -1,4 +1,4 @@
-/*! tailwindcss v4.2.1 | MIT License | https://tailwindcss.com */
+/*! tailwindcss v4.2.0 | MIT License | https://tailwindcss.com */
@layer properties;
@layer theme, base, components, utilities;
@layer theme {
@@ -162,6 +162,59 @@
  }
 }
@layer utilities {
  .diff {
    @layer daisyui.l1.l2.l3 {
      position: relative;
      display: grid;
      width: 100%;
      overflow: hidden;
      webkit-user-select: none;
      user-select: none;
      grid-template-rows: 1fr 1.8rem 1fr;
      direction: ltr;
      container-type: inline-size;
      grid-template-columns: auto 1fr;
      &:focus-visible, &:has(.diff-item-1:focus-visible) {
        outline-style: var(--tw-outline-style);
        outline-width: 2px;
        outline-offset: 1px;
        outline-color: var(--color-base-content);
      }
      &:focus-visible {
        outline-style: var(--tw-outline-style);
        outline-width: 2px;
        outline-offset: 1px;
        outline-color: var(--color-base-content);
        .diff-resizer {
          min-width: 95cqi;
          max-width: 95cqi;
        }
      }
      &:has(.diff-item-1:focus-visible) {
        outline-style: var(--tw-outline-style);
        outline-width: 2px;
        outline-offset: 1px;
        .diff-resizer {
          min-width: 5cqi;
          max-width: 5cqi;
        }
      }
      @supports (-webkit-overflow-scrolling: touch) and (overflow: -webkit-paged-x) {
        &:focus {
          .diff-resizer {
            min-width: 5cqi;
            max-width: 5cqi;
          }
        }
        &:has(.diff-item-1:focus) {
          .diff-resizer {
            min-width: 95cqi;
            max-width: 95cqi;
          }
        }
      }
    }
  }
  .modal {
    @layer daisyui.l1.l2.l3 {
      pointer-events: none;
@@ -1057,98 +1110,31 @@
      }
    }
  }
-  .range {
+  .chat-bubble {
    @layer daisyui.l1.l2.l3 {
-      appearance: none;
+      position: relative;
-      webkit-appearance: none;
+      display: block;
-      --range-thumb: var(--color-base-100);
+      width: fit-content;
-      --range-thumb-size: calc(var(--size-selector, 0.25rem) * 6);
+      border-radius: var(--radius-field);
-      --range-progress: currentColor;
+      background-color: var(--color-base-300);
-      --range-fill: 1;
+      padding-inline: calc(0.25rem * 4);
-      --range-p: 0.25rem;
+      padding-block: calc(0.25rem * 2);
-      --range-bg: currentColor;
+      color: var(--color-base-content);
-      @supports (color: color-mix(in lab, red, red)) {
+      grid-row-end: 3;
-        --range-bg: color-mix(in oklab, currentColor 10%, #0000);
+      min-height: 2rem;
-      }
+      min-width: 2.5rem;
-      cursor: pointer;
+      max-width: 90%;
-      overflow: hidden;
+      &:before {
-      background-color: transparent;
+        position: absolute;
-      vertical-align: middle;
+        bottom: calc(0.25rem * 0);
-      width: clamp(3rem, 20rem, 100%);
+        height: calc(0.25rem * 3);
-      --radius-selector-max: calc(
+        width: calc(0.25rem * 3);
-      var(--radius-selector) + var(--radius-selector) + var(--radius-selector)
+        background-color: inherit;
-    );
+        content: "";
-      border-radius: calc(var(--radius-selector) + min(var(--range-p), var(--radius-selector-max)));
+        mask-repeat: no-repeat;
-      border: none;
+        mask-image: var(--mask-chat);
-      height: var(--range-thumb-size);
+        mask-position: 0px -1px;
-      [dir="rtl"] & {
+        mask-size: 0.8125rem;
        --range-dir: -1;
      }
      &:focus {
        outline: none;
      }
      &:focus-visible {
        outline: 2px solid;
        outline-offset: 2px;
      }
      &::-webkit-slider-runnable-track {
        width: 100%;
        background-color: var(--range-bg);
        border-radius: var(--radius-selector);
        height: calc(var(--range-thumb-size) * 0.5);
      }
      @media (forced-colors: active) {
        &::-webkit-slider-runnable-track {
          border: 1px solid;
        }
      }
      @media (forced-colors: active) {
        &::-moz-range-track {
          border: 1px solid;
        }
      }
      &::-webkit-slider-thumb {
        position: relative;
        box-sizing: border-box;
        border-radius: calc(var(--radius-selector) + min(var(--range-p), var(--radius-selector-max)));
        background-color: var(--range-thumb);
        height: var(--range-thumb-size);
        width: var(--range-thumb-size);
        border: var(--range-p) solid;
        appearance: none;
        webkit-appearance: none;
        top: 50%;
        color: var(--range-progress);
        transform: translateY(-50%);
        box-shadow: 0 -1px oklch(0% 0 0 / calc(var(--depth) * 0.1)) inset, 0 8px 0 -4px oklch(100% 0 0 / calc(var(--depth) * 0.1)) inset, 0 1px currentColor, 0 0 0 2rem var(--range-thumb) inset, calc((var(--range-dir, 1) * -100cqw) - (var(--range-dir, 1) * var(--range-thumb-size) / 2)) 0 0 calc(100cqw * var(--range-fill));
        @supports (color: color-mix(in lab, red, red)) {
          box-shadow: 0 -1px oklch(0% 0 0 / calc(var(--depth) * 0.1)) inset, 0 8px 0 -4px oklch(100% 0 0 / calc(var(--depth) * 0.1)) inset, 0 1px color-mix(in oklab, currentColor calc(var(--depth) * 10%), #0000), 0 0 0 2rem var(--range-thumb) inset, calc((var(--range-dir, 1) * -100cqw) - (var(--range-dir, 1) * var(--range-thumb-size) / 2)) 0 0 calc(100cqw * var(--range-fill));
        }
      }
      &::-moz-range-track {
        width: 100%;
        background-color: var(--range-bg);
        border-radius: var(--radius-selector);
        height: calc(var(--range-thumb-size) * 0.5);
      }
      &::-moz-range-thumb {
        position: relative;
        box-sizing: border-box;
        border-radius: calc(var(--radius-selector) + min(var(--range-p), var(--radius-selector-max)));
        background-color: currentColor;
        height: var(--range-thumb-size);
        width: var(--range-thumb-size);
        border: var(--range-p) solid;
        top: 50%;
        color: var(--range-progress);
        box-shadow: 0 -1px oklch(0% 0 0 / calc(var(--depth) * 0.1)) inset, 0 8px 0 -4px oklch(100% 0 0 / calc(var(--depth) * 0.1)) inset, 0 1px currentColor, 0 0 0 2rem var(--range-thumb) inset, calc((var(--range-dir, 1) * -100cqw) - (var(--range-dir, 1) * var(--range-thumb-size) / 2)) 0 0 calc(100cqw * var(--range-fill));
        @supports (color: color-mix(in lab, red, red)) {
          box-shadow: 0 -1px oklch(0% 0 0 / calc(var(--depth) * 0.1)) inset, 0 8px 0 -4px oklch(100% 0 0 / calc(var(--depth) * 0.1)) inset, 0 1px color-mix(in oklab, currentColor calc(var(--depth) * 10%), #0000), 0 0 0 2rem var(--range-thumb) inset, calc((var(--range-dir, 1) * -100cqw) - (var(--range-dir, 1) * var(--range-thumb-size) / 2)) 0 0 calc(100cqw * var(--range-fill));
        }
      }
      &:disabled {
        cursor: not-allowed;
        opacity: 30%;
      }
    }
  }
@@ -1539,6 +1525,81 @@
      padding: calc(0.25rem * 4);
    }
  }
  .textarea {
    @layer daisyui.l1.l2.l3 {
      border: var(--border) solid #0000;
      min-height: calc(0.25rem * 20);
      flex-shrink: 1;
      appearance: none;
      border-radius: var(--radius-field);
      background-color: var(--color-base-100);
      padding-block: calc(0.25rem * 2);
      vertical-align: middle;
      width: clamp(3rem, 20rem, 100%);
      padding-inline-start: 0.75rem;
      padding-inline-end: 0.75rem;
      font-size: max(var(--font-size, 0.875rem), 0.875rem);
      touch-action: manipulation;
      border-color: var(--input-color);
      box-shadow: 0 1px var(--input-color) inset, 0 -1px oklch(100% 0 0 / calc(var(--depth) * 0.1)) inset;
      @supports (color: color-mix(in lab, red, red)) {
        box-shadow: 0 1px color-mix(in oklab, var(--input-color) calc(var(--depth) * 10%), #0000) inset, 0 -1px oklch(100% 0 0 / calc(var(--depth) * 0.1)) inset;
      }
      --input-color: var(--color-base-content);
      @supports (color: color-mix(in lab, red, red)) {
        --input-color: color-mix(in oklab, var(--color-base-content) 20%, #0000);
      }
      textarea {
        appearance: none;
        background-color: transparent;
        border: none;
        &:focus, &:focus-within {
          --tw-outline-style: none;
          outline-style: none;
          @media (forced-colors: active) {
            outline: 2px solid transparent;
            outline-offset: 2px;
          }
        }
      }
      &:focus, &:focus-within {
        --input-color: var(--color-base-content);
        box-shadow: 0 1px var(--input-color);
        @supports (color: color-mix(in lab, red, red)) {
          box-shadow: 0 1px color-mix(in oklab, var(--input-color) calc(var(--depth) * 10%), #0000);
        }
        outline: 2px solid var(--input-color);
        outline-offset: 2px;
        isolation: isolate;
      }
      @media (pointer: coarse) {
        @supports (-webkit-touch-callout: none) {
          &:focus, &:focus-within {
            --font-size: 1rem;
          }
        }
      }
      &:has(> textarea[disabled]), &:is(:disabled, [disabled]) {
        cursor: not-allowed;
        border-color: var(--color-base-200);
        background-color: var(--color-base-200);
        color: var(--color-base-content);
        @supports (color: color-mix(in lab, red, red)) {
          color: color-mix(in oklab, var(--color-base-content) 40%, transparent);
        }
        &::placeholder {
          color: var(--color-base-content);
          @supports (color: color-mix(in lab, red, red)) {
            color: color-mix(in oklab, var(--color-base-content) 20%, transparent);
          }
        }
        box-shadow: none;
      }
      &:has(> textarea[disabled]) > textarea[disabled] {
        cursor: not-allowed;
      }
    }
  }
  .stack {
    @layer daisyui.l1.l2.l3 {
      display: inline-grid;
--- a/bun.lock
+++ b/bun.lock
@@ -8,7 +8,6 @@
        "tailwindcss": "^4.1.18",
      },
      "devDependencies": {
        "@playwright/test": "^1.52.0",
        "@types/bun": "latest",
      },
      "peerDependencies": {
@@ -17,8 +16,6 @@
    },
  },
  "packages": {
    "@playwright/test": ["@playwright/test@1.58.2", "", { "dependencies": { "playwright": "1.58.2" }, "bin": { "playwright": "cli.js" } }, "sha512-akea+6bHYBBfA9uQqSYmlJXn61cTa+jbO87xVLCWbTqbWadRVmhxlXATaOjOgcBaWU4ePo0wB41KMFv3o35IXA=="],
    "@types/bun": ["@types/bun@1.3.9", "", { "dependencies": { "bun-types": "1.3.9" } }, "sha512-KQ571yULOdWJiMH+RIWIOZ7B2RXQGpL1YQrBtLIV3FqDcCu6FsbFUBwhdKUlCKUpS3PJDsHlJ1QKlpxoVR+xtw=="],
    "@types/node": ["@types/node@25.2.3", "", { "dependencies": { "undici-types": "~7.16.0" } }, "sha512-m0jEgYlYz+mDJZ2+F4v8D1AyQb+QzsNqRuI7xg1VQX/KlKS0qT9r1Mo16yo5F/MtifXFgaofIFsdFMox2SxIbQ=="],
@@ -27,12 +24,6 @@
    "daisyui": ["daisyui@5.5.18", "", {}, "sha512-VVzjpOitMGB6DWIBeRSapbjdOevFqyzpk9u5Um6a4tyId3JFrU5pbtF0vgjXDth76mJZbueN/j9Ok03SPrh/og=="],
    "fsevents": ["fsevents@2.3.2", "", { "os": "darwin" }, "sha512-xiqMQR4xAeHTuB9uWm+fFRcIOgKBMiOBP+eXiyT7jsgVCq1bkVygt00oASowB7EdtpOHaaPgKt812P9ab+DDKA=="],
    "playwright": ["playwright@1.58.2", "", { "dependencies": { "playwright-core": "1.58.2" }, "optionalDependencies": { "fsevents": "2.3.2" }, "bin": { "playwright": "cli.js" } }, "sha512-vA30H8Nvkq/cPBnNw4Q8TWz1EJyqgpuinBcHET0YVJVFldr8JDNiU9LaWAE1KqSkRYazuaBhTpB5ZzShOezQ6A=="],
    "playwright-core": ["playwright-core@1.58.2", "", { "bin": { "playwright-core": "cli.js" } }, "sha512-yZkEtftgwS8CsfYo7nm0KE8jsvm6i/PTgVtB8DL726wNf6H2IMsDuxCpJj59KDaxCtSnrWan2AeDqM7JBaultg=="],
    "tailwindcss": ["tailwindcss@4.1.18", "", {}, "sha512-4+Z+0yiYyEtUVCScyfHCxOYP06L5Ne+JiHhY2IjR2KWMIWhJOYZKLSGZaP5HkZ8+bY0cxfzwDE5uOmzFXyIwxw=="],
    "typescript": ["typescript@5.9.3", "", { "bin": { "tsc": "bin/tsc", "tsserver": "bin/tsserver" } }, "sha512-jl1vZzPDinLr9eUt3J/t7V6FgNEw9QjvBPdysz9KfQDD41fQrC2Y4vKQdiaUpFT4bXlb1RHhLpp8wtm6M5TgSw=="],
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -55,8 +55,6 @@ services:
      mongo:
        condition: service_started
    environment:
      # LiteLLM API key (used by librechat.yaml endpoint config)
      LITELLM_API_KEY: ${LITELLM_API_KEY:-}
      # MongoDB (use localhost since we're on host network)
      MONGO_URI: mongodb://root:example@localhost:27017/librechat?authSource=admin
      DOMAIN_CLIENT: http://localhost:3080
@@ -72,6 +70,7 @@ services:
      OPENID_CALLBACK_URL: /oauth/openid/callback
      OPENID_SCOPE: openid profile email
      OPENID_BUTTON_LABEL: Login with CERTifAI
      OPENID_AUTH_EXTRA_PARAMS: prompt=none
      # Disable local auth (SSO only)
      ALLOW_EMAIL_LOGIN: "false"
      ALLOW_REGISTRATION: "false"
@@ -95,164 +94,5 @@ services:
      - ./librechat/openidStrategy.js:/app/api/strategies/openidStrategy.js:ro
      - librechat-data:/app/data
  langflow:
    image: langflowai/langflow:latest
    container_name: certifai-langflow
    restart: unless-stopped
    ports:
      - "7860:7860"
    environment:
      LANGFLOW_AUTO_LOGIN: "true"
  langgraph:
    image: langchain/langgraph-trial:3.12
    container_name: certifai-langgraph
    restart: unless-stopped
    depends_on:
      langgraph-db:
        condition: service_started
      langgraph-redis:
        condition: service_started
    ports:
      - "8123:8000"
    environment:
      DATABASE_URI: postgresql://langgraph:langgraph@langgraph-db:5432/langgraph
      REDIS_URI: redis://langgraph-redis:6379
  langgraph-db:
    image: postgres:16
    container_name: certifai-langgraph-db
    restart: unless-stopped
    environment:
      POSTGRES_USER: langgraph
      POSTGRES_PASSWORD: langgraph
      POSTGRES_DB: langgraph
    volumes:
      - langgraph-db-data:/var/lib/postgresql/data
  langgraph-redis:
    image: redis:7-alpine
    container_name: certifai-langgraph-redis
    restart: unless-stopped
  langfuse:
    image: langfuse/langfuse:3
    container_name: certifai-langfuse
    restart: unless-stopped
    depends_on:
      keycloak:
        condition: service_healthy
      langfuse-db:
        condition: service_healthy
      langfuse-clickhouse:
        condition: service_healthy
      langfuse-redis:
        condition: service_healthy
      langfuse-minio:
        condition: service_healthy
    ports:
      - "3000:3000"
    environment:
      DATABASE_URL: postgresql://langfuse:langfuse@langfuse-db:5432/langfuse
      NEXTAUTH_URL: http://localhost:3000
      NEXTAUTH_SECRET: certifai-langfuse-dev-secret
      SALT: certifai-langfuse-dev-salt
      ENCRYPTION_KEY: "0000000000000000000000000000000000000000000000000000000000000000"
      # Keycloak OIDC SSO - shared realm with CERTifAI dashboard
      AUTH_KEYCLOAK_CLIENT_ID: certifai-langfuse
      AUTH_KEYCLOAK_CLIENT_SECRET: certifai-langfuse-secret
      AUTH_KEYCLOAK_ISSUER: http://keycloak:8080/realms/certifai
      AUTH_KEYCLOAK_ALLOW_ACCOUNT_LINKING: "true"
      # Disable local email/password auth (SSO only)
      AUTH_DISABLE_USERNAME_PASSWORD: "true"
      CLICKHOUSE_URL: http://langfuse-clickhouse:8123
      CLICKHOUSE_MIGRATION_URL: clickhouse://langfuse-clickhouse:9000
      CLICKHOUSE_USER: clickhouse
      CLICKHOUSE_PASSWORD: clickhouse
      CLICKHOUSE_CLUSTER_ENABLED: "false"
      REDIS_HOST: langfuse-redis
      REDIS_PORT: "6379"
      REDIS_AUTH: langfuse-dev-redis
      LANGFUSE_S3_EVENT_UPLOAD_BUCKET: langfuse
      LANGFUSE_S3_EVENT_UPLOAD_REGION: auto
      LANGFUSE_S3_EVENT_UPLOAD_ACCESS_KEY_ID: minio
      LANGFUSE_S3_EVENT_UPLOAD_SECRET_ACCESS_KEY: miniosecret
      LANGFUSE_S3_EVENT_UPLOAD_ENDPOINT: http://langfuse-minio:9000
      LANGFUSE_S3_EVENT_UPLOAD_FORCE_PATH_STYLE: "true"
      LANGFUSE_S3_MEDIA_UPLOAD_BUCKET: langfuse
      LANGFUSE_S3_MEDIA_UPLOAD_REGION: auto
      LANGFUSE_S3_MEDIA_UPLOAD_ACCESS_KEY_ID: minio
      LANGFUSE_S3_MEDIA_UPLOAD_SECRET_ACCESS_KEY: miniosecret
      LANGFUSE_S3_MEDIA_UPLOAD_ENDPOINT: http://langfuse-minio:9000
      LANGFUSE_S3_MEDIA_UPLOAD_FORCE_PATH_STYLE: "true"
  langfuse-db:
    image: postgres:16
    container_name: certifai-langfuse-db
    restart: unless-stopped
    environment:
      POSTGRES_USER: langfuse
      POSTGRES_PASSWORD: langfuse
      POSTGRES_DB: langfuse
    volumes:
      - langfuse-db-data:/var/lib/postgresql/data
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U langfuse"]
      interval: 5s
      timeout: 5s
      retries: 10
  langfuse-clickhouse:
    image: clickhouse/clickhouse-server:latest
    container_name: certifai-langfuse-clickhouse
    restart: unless-stopped
    user: "101:101"
    environment:
      CLICKHOUSE_DB: default
      CLICKHOUSE_USER: clickhouse
      CLICKHOUSE_PASSWORD: clickhouse
    ulimits:
      nofile:
        soft: 262144
        hard: 262144
    volumes:
      - langfuse-clickhouse-data:/var/lib/clickhouse
      - langfuse-clickhouse-logs:/var/log/clickhouse-server
    healthcheck:
      test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://localhost:8123/ping || exit 1"]
      interval: 5s
      timeout: 5s
      retries: 10
  langfuse-redis:
    image: redis:7-alpine
    container_name: certifai-langfuse-redis
    restart: unless-stopped
    command: redis-server --requirepass langfuse-dev-redis
    healthcheck:
      test: ["CMD", "redis-cli", "-a", "langfuse-dev-redis", "ping"]
      interval: 5s
      timeout: 5s
      retries: 10
  langfuse-minio:
    image: cgr.dev/chainguard/minio
    container_name: certifai-langfuse-minio
    restart: unless-stopped
    entrypoint: sh
    command: -c 'mkdir -p /data/langfuse && minio server --address ":9000" --console-address ":9001" /data'
    environment:
      MINIO_ROOT_USER: minio
      MINIO_ROOT_PASSWORD: miniosecret
    healthcheck:
      test: ["CMD-SHELL", "mc ready local || exit 1"]
      interval: 5s
      timeout: 5s
      retries: 10
 volumes:
  librechat-data:
  langgraph-db-data:
  langfuse-db-data:
  langfuse-clickhouse-data:
  langfuse-clickhouse-logs:
--- a/e2e/auth.setup.ts
+++ b/e2e/auth.setup.ts
@@ -1,24 +0,0 @@
 import { test as setup, expect } from "@playwright/test";
 const AUTH_FILE = "e2e/.auth/user.json";
 setup("authenticate via Keycloak", async ({ page }) => {
  // Navigate to a protected route to trigger the auth redirect chain:
  // /dashboard -> /auth (Axum) -> Keycloak login page
  await page.goto("/dashboard");
  // Wait for Keycloak login form to appear
  await page.waitForSelector("#username", { timeout: 15_000 });
  // Fill Keycloak credentials
  await page.fill("#username", process.env.TEST_USER ?? "admin@certifai.local");
  await page.fill("#password", process.env.TEST_PASSWORD ?? "admin");
  await page.click("#kc-login");
  // Wait for redirect back to the app dashboard
  await page.waitForURL("**/dashboard", { timeout: 15_000 });
  await expect(page.locator(".sidebar")).toBeVisible();
  // Persist authenticated state (cookies + localStorage)
  await page.context().storageState({ path: AUTH_FILE });
 });
--- a/e2e/auth.spec.ts
+++ b/e2e/auth.spec.ts
@@ -1,72 +0,0 @@
 import { test, expect } from "@playwright/test";
 // These tests use a fresh browser context (no saved auth state)
 test.use({ storageState: { cookies: [], origins: [] } });
 test.describe("Authentication flow", () => {
  test("unauthenticated visit to /dashboard redirects to Keycloak", async ({
    page,
  }) => {
    await page.goto("/dashboard");
    // Should end up on Keycloak login page
    await page.waitForSelector("#username", { timeout: 15_000 });
    await expect(page.locator("#kc-login")).toBeVisible();
  });
  test("valid credentials log in and redirect to dashboard", async ({
    page,
  }) => {
    await page.goto("/dashboard");
    await page.waitForSelector("#username", { timeout: 15_000 });
    await page.fill(
      "#username",
      process.env.TEST_USER ?? "admin@certifai.local"
    );
    await page.fill("#password", process.env.TEST_PASSWORD ?? "admin");
    await page.click("#kc-login");
    await page.waitForURL("**/dashboard", { timeout: 15_000 });
    await expect(page.locator(".dashboard-page")).toBeVisible();
  });
  test("dashboard shows sidebar with user info after login", async ({
    page,
  }) => {
    await page.goto("/dashboard");
    await page.waitForSelector("#username", { timeout: 15_000 });
    await page.fill(
      "#username",
      process.env.TEST_USER ?? "admin@certifai.local"
    );
    await page.fill("#password", process.env.TEST_PASSWORD ?? "admin");
    await page.click("#kc-login");
    await page.waitForURL("**/dashboard", { timeout: 15_000 });
    await expect(page.locator(".sidebar-name")).toBeVisible();
    await expect(page.locator(".sidebar-email")).toBeVisible();
  });
  test("logout redirects away from dashboard", async ({ page }) => {
    // First log in
    await page.goto("/dashboard");
    await page.waitForSelector("#username", { timeout: 15_000 });
    await page.fill(
      "#username",
      process.env.TEST_USER ?? "admin@certifai.local"
    );
    await page.fill("#password", process.env.TEST_PASSWORD ?? "admin");
    await page.click("#kc-login");
    await page.waitForURL("**/dashboard", { timeout: 15_000 });
    // Click logout
    await page.locator('a.logout-btn, a[href="/logout"]').click();
    // Should no longer be on the dashboard
    await expect(page).not.toHaveURL(/\/dashboard/);
  });
 });
--- a/e2e/dashboard.spec.ts
+++ b/e2e/dashboard.spec.ts
@@ -1,75 +0,0 @@
 import { test, expect } from "@playwright/test";
 test.describe("Dashboard", () => {
  test.beforeEach(async ({ page }) => {
    await page.goto("/dashboard");
    // Wait for WASM hydration and auth check to complete
    await page.waitForSelector(".dashboard-page", { timeout: 15_000 });
  });
  test("dashboard page loads with page header", async ({ page }) => {
    await expect(page.locator(".page-header")).toContainText("Dashboard");
  });
  test("default topic chips are visible", async ({ page }) => {
    const topics = ["AI", "Technology", "Science", "Finance", "Writing", "Research"];
    for (const topic of topics) {
      await expect(
        page.locator(".filter-tab", { hasText: topic })
      ).toBeVisible();
    }
  });
  test("clicking a topic chip triggers search", async ({ page }) => {
    const chip = page.locator(".filter-tab", { hasText: "AI" });
    await chip.click();
    // Either a loading state or results should appear
    const searchingOrResults = page
      .locator(".dashboard-loading, .news-grid, .dashboard-empty");
    await expect(searchingOrResults.first()).toBeVisible({ timeout: 10_000 });
  });
  test("news cards render after search completes", async ({ page }) => {
    // Click a topic to trigger search
    await page.locator(".filter-tab", { hasText: "Technology" }).click();
    // Wait for loading to finish
    await page.waitForSelector(".dashboard-loading", {
      state: "hidden",
      timeout: 15_000,
    }).catch(() => {
      // Loading may already be done
    });
    // Either news cards or an empty state message should be visible
    const content = page.locator(".news-grid .news-card, .dashboard-empty");
    await expect(content.first()).toBeVisible({ timeout: 10_000 });
  });
  test("clicking a news card opens article detail panel", async ({ page }) => {
    // Trigger a search and wait for results
    await page.locator(".filter-tab", { hasText: "AI" }).click();
    await page.waitForSelector(".dashboard-loading", {
      state: "hidden",
      timeout: 15_000,
    }).catch(() => {});
    const firstCard = page.locator(".news-card").first();
    // Only test if cards are present (search results depend on live data)
    if (await firstCard.isVisible().catch(() => false)) {
      await firstCard.click();
      await expect(page.locator(".dashboard-right, .dashboard-split")).toBeVisible();
    }
  });
  test("settings toggle opens settings panel", async ({ page }) => {
    const settingsBtn = page.locator(".settings-toggle");
    await settingsBtn.click();
    await expect(page.locator(".settings-panel")).toBeVisible();
    await expect(page.locator(".settings-panel-title")).toBeVisible();
  });
 });
--- a/e2e/developer.spec.ts
+++ b/e2e/developer.spec.ts
@@ -1,173 +0,0 @@
 import { test, expect } from "@playwright/test";
 test.describe("Developer section", () => {
  test("agents page loads with sub-nav tabs", async ({ page }) => {
    await page.goto("/developer/agents");
    await page.waitForSelector(".developer-shell", { timeout: 15_000 });
    const nav = page.locator(".sub-nav");
    await expect(nav.locator("a", { hasText: "Agents" })).toBeVisible();
    await expect(nav.locator("a", { hasText: "Flow" })).toBeVisible();
    await expect(nav.locator("a", { hasText: "Analytics" })).toBeVisible();
  });
  test("agents page renders informational landing", async ({ page }) => {
    await page.goto("/developer/agents");
    await page.waitForSelector(".agents-page", { timeout: 15_000 });
    // Hero section
    await expect(page.locator(".agents-hero-title")).toContainText(
      "Agent Builder"
    );
    await expect(page.locator(".agents-hero-desc")).toBeVisible();
    // Connection status indicator is present
    await expect(page.locator(".agents-status")).toBeVisible();
  });
  test("agents page shows Not Connected when URL is empty", async ({
    page,
  }) => {
    await page.goto("/developer/agents");
    await page.waitForSelector(".agents-page", { timeout: 15_000 });
    await expect(page.locator(".agents-status")).toContainText(
      "Not Connected"
    );
    await expect(page.locator(".agents-status-dot--off")).toBeVisible();
    await expect(page.locator(".agents-status-hint")).toBeVisible();
  });
  test("agents page shows quick start cards", async ({ page }) => {
    await page.goto("/developer/agents");
    await page.waitForSelector(".agents-page", { timeout: 15_000 });
    const grid = page.locator(".agents-grid");
    const cards = grid.locator(".agents-card");
    await expect(cards).toHaveCount(5);
    // Verify card titles are rendered
    await expect(
      grid.locator(".agents-card-title", { hasText: "Documentation" })
    ).toBeVisible();
    await expect(
      grid.locator(".agents-card-title", { hasText: "Getting Started" })
    ).toBeVisible();
    await expect(
      grid.locator(".agents-card-title", { hasText: "GitHub" })
    ).toBeVisible();
    await expect(
      grid.locator(".agents-card-title", { hasText: "Examples" })
    ).toBeVisible();
    await expect(
      grid.locator(".agents-card-title", { hasText: "API Reference" })
    ).toBeVisible();
  });
  test("agents page disables API Reference card when not connected", async ({
    page,
  }) => {
    await page.goto("/developer/agents");
    await page.waitForSelector(".agents-page", { timeout: 15_000 });
    // When LANGGRAPH_URL is empty, the API Reference card should be disabled
    const statusHint = page.locator(".agents-status-hint");
    if (await statusHint.isVisible()) {
      const apiCard = page.locator(".agents-card--disabled");
      await expect(apiCard).toBeVisible();
      await expect(
        apiCard.locator(".agents-card-title")
      ).toContainText("API Reference");
    }
  });
  test("agents page shows running agents section", async ({ page }) => {
    await page.goto("/developer/agents");
    await page.waitForSelector(".agents-page", { timeout: 15_000 });
    // The running agents section title should always be visible
    await expect(
      page.locator(".agents-section-title", { hasText: "Running Agents" })
    ).toBeVisible();
    // Either the table, loading state, or empty message should appear
    await page.waitForTimeout(3000);
    const table = page.locator(".agents-table");
    const empty = page.locator(".agents-table-empty");
    const hasTable = await table.isVisible();
    const hasEmpty = await empty.isVisible();
    expect(hasTable || hasEmpty).toBeTruthy();
  });
  test("agents page shows connected status when URL is configured", async ({
    page,
  }) => {
    // This test only passes when LANGGRAPH_URL is set in the environment.
    await page.goto("/developer/agents");
    await page.waitForSelector(".agents-page", { timeout: 15_000 });
    const connectedDot = page.locator(".agents-status-dot--on");
    const disconnectedDot = page.locator(".agents-status-dot--off");
    if (await connectedDot.isVisible()) {
      await expect(page.locator(".agents-status")).toContainText("Connected");
      await expect(page.locator(".agents-status-url")).toBeVisible();
      // API Reference card should NOT be disabled
      await expect(page.locator(".agents-card--disabled")).toHaveCount(0);
    } else {
      await expect(disconnectedDot).toBeVisible();
      await expect(page.locator(".agents-status")).toContainText(
        "Not Connected"
      );
    }
  });
  test("analytics page renders informational landing", async ({ page }) => {
    await page.goto("/developer/analytics");
    await page.waitForSelector(".analytics-page", { timeout: 15_000 });
    // Hero section
    await expect(page.locator(".analytics-hero-title")).toBeVisible();
    await expect(page.locator(".analytics-hero-desc")).toBeVisible();
    // Connection status indicator
    await expect(page.locator(".agents-status")).toBeVisible();
    // Metrics bar
    await expect(page.locator(".analytics-stats-bar")).toBeVisible();
  });
  test("analytics page shows Not Connected when URL is empty", async ({
    page,
  }) => {
    await page.goto("/developer/analytics");
    await page.waitForSelector(".analytics-page", { timeout: 15_000 });
    await expect(page.locator(".agents-status")).toContainText(
      "Not Connected"
    );
    await expect(page.locator(".agents-status-dot--off")).toBeVisible();
  });
  test("analytics page shows quick action cards", async ({ page }) => {
    await page.goto("/developer/analytics");
    await page.waitForSelector(".analytics-page", { timeout: 15_000 });
    const grid = page.locator(".agents-grid");
    const cards = grid.locator(".agents-card, .agents-card--disabled");
    await expect(cards).toHaveCount(2);
  });
  test("analytics page shows SSO hint when connected", async ({ page }) => {
    // Only meaningful when LANGFUSE_URL is configured.
    await page.goto("/developer/analytics");
    await page.waitForSelector(".analytics-page", { timeout: 15_000 });
    const connectedDot = page.locator(".agents-status-dot--on");
    if (await connectedDot.isVisible()) {
      await expect(page.locator(".analytics-sso-hint")).toBeVisible();
      await expect(page.locator(".analytics-launch-btn")).toBeVisible();
    }
  });
 });
--- a/e2e/navigation.spec.ts
+++ b/e2e/navigation.spec.ts
@@ -1,52 +0,0 @@
 import { test, expect } from "@playwright/test";
 test.describe("Sidebar navigation", () => {
  test.beforeEach(async ({ page }) => {
    await page.goto("/dashboard");
    await page.waitForSelector(".sidebar", { timeout: 15_000 });
  });
  test("sidebar links route to correct pages", async ({ page }) => {
    const navTests = [
      { label: "Providers", url: /\/providers/ },
      { label: "Developer", url: /\/developer\/agents/ },
      { label: "Organization", url: /\/organization\/pricing/ },
      { label: "Dashboard", url: /\/dashboard/ },
    ];
    for (const { label, url } of navTests) {
      await page.locator(".sidebar-link", { hasText: label }).click();
      await expect(page).toHaveURL(url, { timeout: 10_000 });
    }
  });
  test("browser back/forward navigation works", async ({ page }) => {
    // Navigate to Providers
    await page.locator(".sidebar-link", { hasText: "Providers" }).click();
    await expect(page).toHaveURL(/\/providers/);
    // Navigate to Developer
    await page.locator(".sidebar-link", { hasText: "Developer" }).click();
    await expect(page).toHaveURL(/\/developer/);
    // Go back
    await page.goBack();
    await expect(page).toHaveURL(/\/providers/);
    // Go forward
    await page.goForward();
    await expect(page).toHaveURL(/\/developer/);
  });
  test("logo link navigates to dashboard", async ({ page }) => {
    // Navigate away first
    await page.locator(".sidebar-link", { hasText: "Providers" }).click();
    await expect(page).toHaveURL(/\/providers/);
    // Click the logo/brand in sidebar header
    const logo = page.locator(".sidebar-brand, .sidebar-logo, .sidebar a").first();
    await logo.click();
    await expect(page).toHaveURL(/\/dashboard/);
  });
 });
--- a/e2e/organization.spec.ts
+++ b/e2e/organization.spec.ts
@@ -1,41 +0,0 @@
 import { test, expect } from "@playwright/test";
 test.describe("Organization section", () => {
  test("pricing page loads with three pricing cards", async ({ page }) => {
    await page.goto("/organization/pricing");
    await page.waitForSelector(".org-shell", { timeout: 15_000 });
    const cards = page.locator(".pricing-card");
    await expect(cards).toHaveCount(3);
  });
  test("pricing cards show Starter, Team, Enterprise tiers", async ({
    page,
  }) => {
    await page.goto("/organization/pricing");
    await page.waitForSelector(".org-shell", { timeout: 15_000 });
    await expect(page.locator(".pricing-card", { hasText: "Starter" })).toBeVisible();
    await expect(page.locator(".pricing-card", { hasText: "Team" })).toBeVisible();
    await expect(page.locator(".pricing-card", { hasText: "Enterprise" })).toBeVisible();
  });
  test("organization dashboard loads with billing stats", async ({ page }) => {
    await page.goto("/organization/dashboard");
    await page.waitForSelector(".org-dashboard-page", { timeout: 15_000 });
    await expect(page.locator(".page-header")).toContainText("Organization");
    await expect(page.locator(".org-stats-bar")).toBeVisible();
    await expect(page.locator(".org-stat").first()).toBeVisible();
  });
  test("member table is visible on org dashboard", async ({ page }) => {
    await page.goto("/organization/dashboard");
    await page.waitForSelector(".org-dashboard-page", { timeout: 15_000 });
    await expect(page.locator(".org-table")).toBeVisible();
    await expect(page.locator(".org-table thead")).toContainText("Name");
    await expect(page.locator(".org-table thead")).toContainText("Email");
    await expect(page.locator(".org-table thead")).toContainText("Role");
  });
 });
--- a/e2e/providers.spec.ts
+++ b/e2e/providers.spec.ts
@@ -1,55 +0,0 @@
 import { test, expect } from "@playwright/test";
 test.describe("Providers page", () => {
  test.beforeEach(async ({ page }) => {
    await page.goto("/providers");
    await page.waitForSelector(".providers-page", { timeout: 15_000 });
  });
  test("providers page loads with header", async ({ page }) => {
    await expect(page.locator(".page-header")).toContainText("Providers");
  });
  test("provider dropdown has Ollama selected by default", async ({
    page,
  }) => {
    const providerSelect = page
      .locator(".form-group")
      .filter({ hasText: "Provider" })
      .locator("select");
    await expect(providerSelect).toHaveValue(/ollama/i);
  });
  test("changing provider updates the model dropdown", async ({ page }) => {
    const providerSelect = page
      .locator(".form-group")
      .filter({ hasText: "Provider" })
      .locator("select");
    // Get current model options
    const modelSelect = page
      .locator(".form-group")
      .filter({ hasText: /^Model/ })
      .locator("select");
    const initialOptions = await modelSelect.locator("option").allTextContents();
    // Change to a different provider
    await providerSelect.selectOption({ label: "OpenAI" });
    // Wait for model list to update
    await page.waitForTimeout(500);
    const updatedOptions = await modelSelect.locator("option").allTextContents();
    // Model options should differ between providers
    expect(updatedOptions).not.toEqual(initialOptions);
  });
  test("save button shows confirmation feedback", async ({ page }) => {
    const saveBtn = page.locator("button", { hasText: "Save Configuration" });
    await saveBtn.click();
    await expect(page.locator(".form-success")).toBeVisible({ timeout: 5_000 });
    await expect(page.locator(".form-success")).toContainText("saved");
  });
 });
--- a/e2e/public.spec.ts
+++ b/e2e/public.spec.ts
@@ -1,60 +0,0 @@
 import { test, expect } from "@playwright/test";
 test.describe("Public pages", () => {
  test("landing page loads with heading and nav links", async ({ page }) => {
    await page.goto("/");
    await expect(page.locator(".landing-logo").first()).toHaveText("CERTifAI");
    await expect(page.locator(".landing-nav-links")).toBeVisible();
    await expect(page.locator('a[href="#features"]')).toBeVisible();
    await expect(page.locator('a[href="#how-it-works"]')).toBeVisible();
    await expect(page.locator('a[href="#pricing"]')).toBeVisible();
  });
  test("landing page Log In link navigates to login route", async ({
    page,
  }) => {
    await page.goto("/");
    const loginLink = page
      .locator(".landing-nav-actions a, .landing-nav-actions Link")
      .filter({ hasText: "Log In" });
    await loginLink.click();
    await expect(page).toHaveURL(/\/login/);
  });
  test("impressum page loads with legal content", async ({ page }) => {
    await page.goto("/impressum");
    await expect(page.locator("h1")).toHaveText("Impressum");
    await expect(
      page.locator("h2", { hasText: "Information according to" })
    ).toBeVisible();
    await expect(page.locator(".legal-content")).toContainText(
      "CERTifAI GmbH"
    );
  });
  test("privacy page loads with privacy content", async ({ page }) => {
    await page.goto("/privacy");
    await expect(page.locator("h1")).toHaveText("Privacy Policy");
    await expect(
      page.locator("h2", { hasText: "Introduction" })
    ).toBeVisible();
    await expect(
      page.locator("h2", { hasText: "Your Rights" })
    ).toBeVisible();
  });
  test("footer links are present on landing page", async ({ page }) => {
    await page.goto("/");
    const footer = page.locator(".landing-footer");
    await expect(footer.locator('a:has-text("Impressum")')).toBeVisible();
    await expect(
      footer.locator('a:has-text("Privacy Policy")')
    ).toBeVisible();
  });
 });
--- a/keycloak/realm-export.json
+++ b/keycloak/realm-export.json
@@ -79,39 +79,6 @@
        "offline_access"
      ]
    },
    {
      "clientId": "certifai-langfuse",
      "name": "CERTifAI Langfuse",
      "description": "Langfuse OIDC client for CERTifAI",
      "enabled": true,
      "publicClient": false,
      "directAccessGrantsEnabled": false,
      "standardFlowEnabled": true,
      "implicitFlowEnabled": false,
      "serviceAccountsEnabled": false,
      "protocol": "openid-connect",
      "secret": "certifai-langfuse-secret",
      "rootUrl": "http://localhost:3000",
      "baseUrl": "http://localhost:3000",
      "redirectUris": [
        "http://localhost:3000/*"
      ],
      "webOrigins": [
        "http://localhost:3000",
        "http://localhost:8000"
      ],
      "attributes": {
        "post.logout.redirect.uris": "http://localhost:3000"
      },
      "defaultClientScopes": [
        "openid",
        "profile",
        "email"
      ],
      "optionalClientScopes": [
        "offline_access"
      ]
    },
    {
      "clientId": "certifai-librechat",
      "name": "CERTifAI Chat",
--- a/librechat/librechat.yaml
+++ b/librechat/librechat.yaml
@@ -1,5 +1,5 @@
 # CERTifAI LibreChat Configuration
-# LiteLLM proxy for unified multi-provider LLM access.
+# Ollama backend for self-hosted LLM inference.
 version: 1.2.8
 cache: true
@@ -19,16 +19,22 @@ interface:
 endpoints:
  custom:
-    - name: "LiteLLM"
+    - name: "Ollama"
-      apiKey: "${LITELLM_API_KEY}"
+      apiKey: "ollama"
-      baseURL: "https://llm-dev.meghsakha.com/v1/"
+      baseURL: "https://mac-mini-von-benjamin-2:11434/v1/"
      models:
        default:
-          - "Qwen3-Coder-30B-A3B-Instruct"
+          - "llama3.1:8b"
          - "qwen3:30b-a3b"
        fetch: true
      titleConvo: true
      titleModel: "current_model"
      summarize: false
      summaryModel: "current_model"
      forcePrompt: false
-      modelDisplayLabel: "CERTifAI LiteLLM"
+      modelDisplayLabel: "CERTifAI Ollama"
      dropParams:
        - stop
        - user
        - frequency_penalty
        - presence_penalty
--- a/package.json
+++ b/package.json
@@ -4,7 +4,6 @@
  "type": "module",
  "private": true,
  "devDependencies": {
    "@playwright/test": "^1.52.0",
    "@types/bun": "latest"
  },
  "peerDependencies": {
--- a/playwright.config.ts
+++ b/playwright.config.ts
@@ -1,40 +0,0 @@
 import { defineConfig, devices } from "@playwright/test";
 export default defineConfig({
  testDir: "./e2e",
  fullyParallel: true,
  forbidOnly: !!process.env.CI,
  retries: process.env.CI ? 2 : 0,
  workers: process.env.CI ? 1 : undefined,
  reporter: [["html"], ["list"]],
  timeout: 30_000,
  use: {
    baseURL: process.env.BASE_URL ?? "http://localhost:8000",
    actionTimeout: 10_000,
    trace: "on-first-retry",
    screenshot: "only-on-failure",
  },
  projects: [
    {
      name: "setup",
      testMatch: /auth\.setup\.ts/,
    },
    {
      name: "public",
      testMatch: /public\.spec\.ts/,
      use: { ...devices["Desktop Chrome"] },
    },
    {
      name: "authenticated",
      testMatch: /\.spec\.ts$/,
      testIgnore: /public\.spec\.ts$/,
      dependencies: ["setup"],
      use: {
        ...devices["Desktop Chrome"],
        storageState: "e2e/.auth/user.json",
      },
    },
  ],
 });
--- a/src/components/app_shell.rs
+++ b/src/components/app_shell.rs
@@ -5,7 +5,7 @@ use dioxus_free_icons::Icon;
 use crate::components::sidebar::Sidebar;
 use crate::i18n::{t, tw, Locale};
 use crate::infrastructure::auth_check::check_auth;
-use crate::models::{AuthInfo, ServiceUrlsContext};
+use crate::models::AuthInfo;
 use crate::Route;
 /// Application shell layout that wraps all authenticated pages.
@@ -29,16 +29,6 @@ pub fn AppShell() -> Element {
    match auth_snapshot {
        Some(Ok(info)) if info.authenticated => {
            // Provide developer tool URLs as context so child pages
            // can read them without prop-drilling through layouts.
            use_context_provider(|| {
                Signal::new(ServiceUrlsContext {
                    langgraph_url: info.langgraph_url.clone(),
                    langflow_url: info.langflow_url.clone(),
                    langfuse_url: info.langfuse_url.clone(),
                })
            });
            let menu_open = *mobile_menu_open.read();
            let sidebar_cls = if menu_open {
                "sidebar sidebar--open"
--- a/src/components/dashboard_sidebar.rs
+++ b/src/components/dashboard_sidebar.rs
@@ -1,9 +1,9 @@
 use dioxus::prelude::*;
 use crate::i18n::{t, Locale};
-use crate::infrastructure::litellm::{get_litellm_status, LitellmStatus};
+use crate::infrastructure::ollama::{get_ollama_status, OllamaStatus};
-/// Right sidebar for the dashboard, showing LiteLLM status, trending topics,
+/// Right sidebar for the dashboard, showing Ollama status, trending topics,
 /// and recent search history.
 ///
 /// Appears when no article card is selected. Disappears when the user opens
@@ -11,13 +11,13 @@ use crate::infrastructure::litellm::{get_litellm_status, LitellmStatus};
 ///
 /// # Props
 ///
-/// * `litellm_url` - LiteLLM proxy URL for status polling
+/// * `ollama_url` - Ollama instance URL for status polling
 /// * `trending` - Trending topic keywords extracted from recent news headlines
 /// * `recent_searches` - Recent search topics stored in localStorage
 /// * `on_topic_click` - Fires when a trending or recent topic is clicked
 #[component]
 pub fn DashboardSidebar(
-    litellm_url: String,
+    ollama_url: String,
    trending: Vec<String>,
    recent_searches: Vec<String>,
    on_topic_click: EventHandler<String>,
@@ -25,26 +25,26 @@ pub fn DashboardSidebar(
    let locale = use_context::<Signal<Locale>>();
    let l = *locale.read();
-    // Fetch LiteLLM status once on mount.
+    // Fetch Ollama status once on mount.
    // use_resource with no signal dependencies runs exactly once and
    // won't re-fire on parent re-renders (unlike use_effect).
-    let url = litellm_url.clone();
+    let url = ollama_url.clone();
    let status_resource = use_resource(move || {
        let u = url.clone();
        async move {
-            get_litellm_status(u).await.unwrap_or(LitellmStatus {
+            get_ollama_status(u).await.unwrap_or(OllamaStatus {
                online: false,
                models: Vec::new(),
            })
        }
    });
-    let current_status: LitellmStatus =
+    let current_status: OllamaStatus =
        status_resource
            .read()
            .as_ref()
            .cloned()
-            .unwrap_or(LitellmStatus {
+            .unwrap_or(OllamaStatus {
                online: false,
                models: Vec::new(),
            });
@@ -52,9 +52,9 @@ pub fn DashboardSidebar(
    rsx! {
        aside { class: "dashboard-sidebar",
-            // -- LiteLLM Status Section --
+            // -- Ollama Status Section --
            div { class: "sidebar-section",
-                h4 { class: "sidebar-section-title", "{t(l, \"dashboard.litellm_status\")}" }
+                h4 { class: "sidebar-section-title", "{t(l, \"dashboard.ollama_status\")}" }
                div { class: "sidebar-status-row",
                    span { class: if current_status.online { "sidebar-status-dot sidebar-status-dot--online" } else { "sidebar-status-dot sidebar-status-dot--offline" } }
                    span { class: "sidebar-status-label",
--- a/src/components/mod.rs
+++ b/src/components/mod.rs
@@ -9,7 +9,6 @@ mod page_header;
 mod pricing_card;
 pub mod sidebar;
 pub mod sub_nav;
 mod tool_embed;
 pub use app_shell::*;
 pub use article_detail::*;
@@ -21,4 +20,3 @@ pub use news_card::*;
 pub use page_header::*;
 pub use pricing_card::*;
 pub use sub_nav::*;
 pub use tool_embed::*;
--- a/src/components/news_card.rs
+++ b/src/components/news_card.rs
@@ -112,12 +112,12 @@ pub fn mock_news() -> Vec<NewsCardModel> {
            published_at: "2026-02-16".into(),
        },
        NewsCardModel {
-            title: "LiteLLM Adds Multi-Provider Routing".into(),
+            title: "Ollama Adds Multi-GPU Scheduling".into(),
-            source: "LiteLLM".into(),
+            source: "Ollama".into(),
-            summary: "Route requests across multiple LLM providers with automatic fallback.".into(),
+            summary: "Run large models across multiple GPUs with automatic sharding.".into(),
-            content: "LiteLLM now supports multi-provider routing with automatic \
+            content: "Ollama now supports multi-GPU scheduling with automatic \
-                fallback. Users can route requests across multiple providers \
+                model sharding. Users can run models across multiple GPUs \
-                for improved reliability and cost optimization."
+                for improved inference performance."
                .into(),
            category: "Infrastructure".into(),
            url: "#".into(),
--- a/src/components/tool_embed.rs
+++ b/src/components/tool_embed.rs
@@ -1,81 +0,0 @@
 use dioxus::prelude::*;
 use crate::i18n::{t, Locale};
 /// Properties for the [`ToolEmbed`] component.
 ///
 /// # Fields
 ///
 /// * `url` - Service URL; when empty, a "Not Configured" placeholder is shown
 /// * `title` - Display title for the tool (e.g. "Agent Builder")
 /// * `description` - Description text shown in the placeholder card
 /// * `icon` - Single-character icon for the placeholder card
 /// * `launch_label` - Label for the disabled button in the placeholder
 #[derive(Props, Clone, PartialEq)]
 pub struct ToolEmbedProps {
    /// Service URL. Empty string means "not configured".
    pub url: String,
    /// Display title shown in the toolbar / placeholder heading.
    pub title: String,
    /// Description shown in the "not configured" placeholder.
    pub description: String,
    /// Single-character icon for the placeholder card.
    pub icon: &'static str,
    /// Label for the disabled launch button in placeholder mode.
    pub launch_label: String,
 }
 /// Hybrid iframe / placeholder component for developer tool pages.
 ///
 /// When `url` is non-empty, renders a toolbar (title + pop-out button)
 /// above a full-height iframe embedding the service. When `url` is
 /// empty, renders the existing placeholder card with a "Not Configured"
 /// badge instead of "Coming Soon".
 #[component]
 pub fn ToolEmbed(props: ToolEmbedProps) -> Element {
    let locale = use_context::<Signal<Locale>>();
    let l = *locale.read();
    if props.url.is_empty() {
        // Not configured -- show placeholder card
        rsx! {
            section { class: "placeholder-page",
                div { class: "placeholder-card",
                    div { class: "placeholder-icon", "{props.icon}" }
                    h2 { "{props.title}" }
                    p { class: "placeholder-desc", "{props.description}" }
                    button {
                        class: "btn-primary",
                        disabled: true,
                        "{props.launch_label}"
                    }
                    span { class: "placeholder-badge",
                        "{t(l, \"developer.not_configured\")}"
                    }
                }
            }
        }
    } else {
        // URL is set -- render toolbar + iframe
        let pop_out_url = props.url.clone();
        rsx! {
            div { class: "tool-embed",
                div { class: "tool-embed-toolbar",
                    span { class: "tool-embed-title", "{props.title}" }
                    a {
                        class: "tool-embed-popout-btn",
                        href: "{pop_out_url}",
                        target: "_blank",
                        rel: "noopener noreferrer",
                        "{t(l, \"developer.open_new_tab\")}"
                    }
                }
                iframe {
                    class: "tool-embed-iframe",
                    src: "{props.url}",
                    title: "{props.title}",
                }
            }
        }
    }
 }
--- a/src/infrastructure/auth.rs
+++ b/src/infrastructure/auth.rs
@@ -24,9 +24,9 @@ pub const LOGGED_IN_USER_SESS_KEY: &str = "logged-in-user";
 /// post-login redirect URL and the PKCE code verifier needed for the
 /// token exchange.
 #[derive(Debug, Clone)]
-pub(crate) struct PendingOAuthEntry {
+struct PendingOAuthEntry {
-    pub(crate) redirect_url: Option<String>,
+    redirect_url: Option<String>,
-    pub(crate) code_verifier: String,
+    code_verifier: String,
 }
 /// In-memory store for pending OAuth states. Keyed by the random state
@@ -38,7 +38,7 @@ pub struct PendingOAuthStore(Arc<RwLock<HashMap<String, PendingOAuthEntry>>>);
 impl PendingOAuthStore {
    /// Insert a pending state with an optional redirect URL and PKCE verifier.
-    pub(crate) fn insert(&self, state: String, entry: PendingOAuthEntry) {
+    fn insert(&self, state: String, entry: PendingOAuthEntry) {
        // RwLock::write only panics if the lock is poisoned, which
        // indicates a prior panic -- propagating is acceptable here.
        #[allow(clippy::expect_used)]
@@ -50,7 +50,7 @@ impl PendingOAuthStore {
    /// Remove and return the entry if the state was pending.
    /// Returns `None` if the state was never stored (CSRF failure).
-    pub(crate) fn take(&self, state: &str) -> Option<PendingOAuthEntry> {
+    fn take(&self, state: &str) -> Option<PendingOAuthEntry> {
        #[allow(clippy::expect_used)]
        self.0
            .write()
@@ -60,8 +60,7 @@ impl PendingOAuthStore {
 }
 /// Generate a cryptographically random state string for CSRF protection.
-#[cfg_attr(test, allow(dead_code))]
+fn generate_state() -> String {
 pub(crate) fn generate_state() -> String {
    let bytes: [u8; 32] = rand::rng().random();
    // Encode as hex to produce a URL-safe string without padding.
    bytes.iter().fold(String::with_capacity(64), |mut acc, b| {
@@ -76,7 +75,7 @@ pub(crate) fn generate_state() -> String {
 ///
 /// Uses 32 random bytes encoded as base64url (no padding) to produce
 /// a 43-character verifier per RFC 7636.
-pub(crate) fn generate_code_verifier() -> String {
+fn generate_code_verifier() -> String {
    use base64::{engine::general_purpose::URL_SAFE_NO_PAD, Engine};
    let bytes: [u8; 32] = rand::rng().random();
@@ -86,7 +85,7 @@ pub(crate) fn generate_code_verifier() -> String {
 /// Derive the S256 code challenge from a code verifier per RFC 7636.
 ///
 /// `code_challenge = BASE64URL(SHA256(code_verifier))`
-pub(crate) fn derive_code_challenge(verifier: &str) -> String {
+fn derive_code_challenge(verifier: &str) -> String {
    use base64::{engine::general_purpose::URL_SAFE_NO_PAD, Engine};
    use sha2::{Digest, Sha256};
@@ -305,117 +304,3 @@ pub async fn set_login_session(session: Session, data: UserStateInner) -> Result
        .await
        .map_err(|e| Error::StateError(format!("session insert failed: {e}")))
 }
 #[cfg(test)]
 mod tests {
    #![allow(clippy::unwrap_used, clippy::expect_used)]
    use super::*;
    use pretty_assertions::assert_eq;
    // -----------------------------------------------------------------------
    // generate_state()
    // -----------------------------------------------------------------------
    #[test]
    fn generate_state_length_is_64() {
        let state = generate_state();
        assert_eq!(state.len(), 64);
    }
    #[test]
    fn generate_state_chars_are_hex() {
        let state = generate_state();
        assert!(state.chars().all(|c| c.is_ascii_hexdigit()));
    }
    #[test]
    fn generate_state_two_calls_differ() {
        let a = generate_state();
        let b = generate_state();
        assert_ne!(a, b);
    }
    // -----------------------------------------------------------------------
    // generate_code_verifier()
    // -----------------------------------------------------------------------
    #[test]
    fn code_verifier_length_is_43() {
        let verifier = generate_code_verifier();
        assert_eq!(verifier.len(), 43);
    }
    #[test]
    fn code_verifier_chars_are_url_safe_base64() {
        let verifier = generate_code_verifier();
        // URL-safe base64 without padding uses [A-Za-z0-9_-]
        assert!(verifier
            .chars()
            .all(|c| c.is_ascii_alphanumeric() || c == '-' || c == '_'));
    }
    // -----------------------------------------------------------------------
    // derive_code_challenge()
    // -----------------------------------------------------------------------
    #[test]
    fn code_challenge_deterministic() {
        let verifier = "dBjftJeZ4CVP-mB92K27uhbUJU1p1r_wW1gFWFOEjXk";
        let a = derive_code_challenge(verifier);
        let b = derive_code_challenge(verifier);
        assert_eq!(a, b);
    }
    #[test]
    fn code_challenge_rfc7636_test_vector() {
        // RFC 7636 Appendix B test vector:
        // verifier = "dBjftJeZ4CVP-mB92K27uhbUJU1p1r_wW1gFWFOEjXk"
        // expected challenge = "E9Melhoa2OwvFrEMTJguCHaoeK1t8URWbuGJSstw-cM"
        let verifier = "dBjftJeZ4CVP-mB92K27uhbUJU1p1r_wW1gFWFOEjXk";
        let challenge = derive_code_challenge(verifier);
        assert_eq!(challenge, "E9Melhoa2OwvFrEMTJguCHaoeK1t8URWbuGJSstw-cM");
    }
    // -----------------------------------------------------------------------
    // PendingOAuthStore
    // -----------------------------------------------------------------------
    #[test]
    fn pending_store_insert_and_take() {
        let store = PendingOAuthStore::default();
        store.insert(
            "state-1".into(),
            PendingOAuthEntry {
                redirect_url: Some("/dashboard".into()),
                code_verifier: "verifier-1".into(),
            },
        );
        let entry = store.take("state-1");
        assert!(entry.is_some());
        let entry = entry.unwrap();
        assert_eq!(entry.redirect_url, Some("/dashboard".into()));
        assert_eq!(entry.code_verifier, "verifier-1");
    }
    #[test]
    fn pending_store_take_removes_entry() {
        let store = PendingOAuthStore::default();
        store.insert(
            "state-2".into(),
            PendingOAuthEntry {
                redirect_url: None,
                code_verifier: "v2".into(),
            },
        );
        let _ = store.take("state-2");
        // Second take should return None since the entry was removed.
        assert!(store.take("state-2").is_none());
    }
    #[test]
    fn pending_store_take_unknown_returns_none() {
        let store = PendingOAuthStore::default();
        assert!(store.take("nonexistent").is_none());
    }
 }
--- a/src/infrastructure/auth_check.rs
+++ b/src/infrastructure/auth_check.rs
@@ -25,17 +25,8 @@ pub async fn check_auth() -> Result<AuthInfo, ServerFnError> {
    match user_state {
        Some(u) => {
-            let librechat_url =
+            let librechat_url = std::env::var("LIBRECHAT_URL")
-                std::env::var("LIBRECHAT_URL").unwrap_or_else(|_| "http://localhost:3080".into());
+                .unwrap_or_else(|_| "http://localhost:3080".into());
            // Extract service URLs from server state so the frontend can
            // embed developer tools (LangGraph, LangFlow, Langfuse).
            let state: crate::infrastructure::server_state::ServerState =
                FullstackContext::extract().await?;
            let langgraph_url = state.services.langgraph_url.clone();
            let langflow_url = state.services.langflow_url.clone();
            let langfuse_url = state.services.langfuse_url.clone();
            Ok(AuthInfo {
                authenticated: true,
                sub: u.sub,
@@ -43,9 +34,6 @@ pub async fn check_auth() -> Result<AuthInfo, ServerFnError> {
                name: u.user.name,
                avatar_url: u.user.avatar_url,
                librechat_url,
                langgraph_url,
                langflow_url,
                langfuse_url,
            })
        }
        None => Ok(AuthInfo::default()),
--- a/src/infrastructure/chat.rs
+++ b/src/infrastructure/chat.rs
@@ -134,7 +134,7 @@ pub async fn list_chat_sessions() -> Result<Vec<ChatSession>, ServerFnError> {
 ///
 /// * `title` - Display title for the session
 /// * `namespace` - Namespace string: `"General"` or `"News"`
-/// * `provider` - LLM provider name (e.g. "litellm")
+/// * `provider` - LLM provider name (e.g. "ollama")
 /// * `model` - Model ID (e.g. "llama3.1:8b")
 /// * `article_url` - Source article URL (only for `News` namespace, empty if none)
 ///
@@ -440,12 +440,7 @@ pub async fn chat_complete(
    let session = doc_to_chat_session(&session_doc);
    // Resolve provider URL and model
-    let (base_url, model) = resolve_provider_url(
+    let (base_url, model) = resolve_provider_url(&state, &session.provider, &session.model);
        &state.services.litellm_url,
        &state.services.litellm_model,
        &session.provider,
        &session.model,
    );
    // Parse messages from JSON
    let chat_msgs: Vec<serde_json::Value> = serde_json::from_str(&messages_json)
@@ -485,22 +480,10 @@ pub async fn chat_complete(
        .ok_or_else(|| ServerFnError::new("empty LLM response"))
 }
-/// Resolve the base URL for a provider, falling back to LiteLLM defaults.
+/// Resolve the base URL for a provider, falling back to server defaults.
 ///
 /// # Arguments
 ///
 /// * `litellm_url` - Default LiteLLM base URL from config
 /// * `litellm_model` - Default LiteLLM model from config
 /// * `provider` - Provider name (e.g. "openai", "anthropic", "huggingface")
 /// * `model` - Model ID (may be empty for LiteLLM default)
 ///
 /// # Returns
 ///
 /// A `(base_url, model)` tuple resolved for the given provider.
 #[cfg(feature = "server")]
-pub(crate) fn resolve_provider_url(
+fn resolve_provider_url(
-    litellm_url: &str,
+    state: &crate::infrastructure::ServerState,
    litellm_model: &str,
    provider: &str,
    model: &str,
 ) -> (String, String) {
@@ -511,231 +494,14 @@ pub(crate) fn resolve_provider_url(
            format!("https://api-inference.huggingface.co/models/{}", model),
            model.to_string(),
        ),
-        // Default to LiteLLM
+        // Default to Ollama
        _ => (
-            litellm_url.to_string(),
+            state.services.ollama_url.clone(),
            if model.is_empty() {
-                litellm_model.to_string()
+                state.services.ollama_model.clone()
            } else {
                model.to_string()
            },
        ),
    }
 }
 #[cfg(test)]
 mod tests {
    // -----------------------------------------------------------------------
    // BSON document conversion tests (server feature required)
    // -----------------------------------------------------------------------
    #[cfg(feature = "server")]
    mod server_tests {
        use super::super::{doc_to_chat_message, doc_to_chat_session, resolve_provider_url};
        use crate::models::{ChatNamespace, ChatRole};
        use mongodb::bson::{doc, oid::ObjectId, Document};
        use pretty_assertions::assert_eq;
        // -- doc_to_chat_session --
        fn sample_session_doc() -> (ObjectId, Document) {
            let oid = ObjectId::new();
            let doc = doc! {
                "_id": oid,
                "user_sub": "user-42",
                "title": "Test Session",
                "namespace": "News",
                "provider": "openai",
                "model": "gpt-4",
                "created_at": "2025-01-01T00:00:00Z",
                "updated_at": "2025-01-02T00:00:00Z",
                "article_url": "https://example.com/article",
            };
            (oid, doc)
        }
        #[test]
        fn doc_to_chat_session_extracts_id_as_hex() {
            let (oid, doc) = sample_session_doc();
            let session = doc_to_chat_session(&doc);
            assert_eq!(session.id, oid.to_hex());
        }
        #[test]
        fn doc_to_chat_session_maps_news_namespace() {
            let (_, doc) = sample_session_doc();
            let session = doc_to_chat_session(&doc);
            assert_eq!(session.namespace, ChatNamespace::News);
        }
        #[test]
        fn doc_to_chat_session_defaults_to_general_for_unknown() {
            let mut doc = sample_session_doc().1;
            doc.insert("namespace", "SomethingElse");
            let session = doc_to_chat_session(&doc);
            assert_eq!(session.namespace, ChatNamespace::General);
        }
        #[test]
        fn doc_to_chat_session_extracts_all_string_fields() {
            let (_, doc) = sample_session_doc();
            let session = doc_to_chat_session(&doc);
            assert_eq!(session.user_sub, "user-42");
            assert_eq!(session.title, "Test Session");
            assert_eq!(session.provider, "openai");
            assert_eq!(session.model, "gpt-4");
            assert_eq!(session.created_at, "2025-01-01T00:00:00Z");
            assert_eq!(session.updated_at, "2025-01-02T00:00:00Z");
        }
        #[test]
        fn doc_to_chat_session_handles_missing_article_url() {
            let oid = ObjectId::new();
            let doc = doc! {
                "_id": oid,
                "user_sub": "u",
                "title": "t",
                "provider": "litellm",
                "model": "m",
                "created_at": "c",
                "updated_at": "u",
            };
            let session = doc_to_chat_session(&doc);
            assert_eq!(session.article_url, None);
        }
        #[test]
        fn doc_to_chat_session_filters_empty_article_url() {
            let oid = ObjectId::new();
            let doc = doc! {
                "_id": oid,
                "user_sub": "u",
                "title": "t",
                "namespace": "News",
                "provider": "litellm",
                "model": "m",
                "created_at": "c",
                "updated_at": "u",
                "article_url": "",
            };
            let session = doc_to_chat_session(&doc);
            assert_eq!(session.article_url, None);
        }
        // -- doc_to_chat_message --
        fn sample_message_doc() -> (ObjectId, Document) {
            let oid = ObjectId::new();
            let doc = doc! {
                "_id": oid,
                "session_id": "sess-1",
                "role": "Assistant",
                "content": "Hello there!",
                "timestamp": "2025-01-01T12:00:00Z",
            };
            (oid, doc)
        }
        #[test]
        fn doc_to_chat_message_extracts_id_as_hex() {
            let (oid, doc) = sample_message_doc();
            let msg = doc_to_chat_message(&doc);
            assert_eq!(msg.id, oid.to_hex());
        }
        #[test]
        fn doc_to_chat_message_maps_assistant_role() {
            let (_, doc) = sample_message_doc();
            let msg = doc_to_chat_message(&doc);
            assert_eq!(msg.role, ChatRole::Assistant);
        }
        #[test]
        fn doc_to_chat_message_maps_system_role() {
            let mut doc = sample_message_doc().1;
            doc.insert("role", "System");
            let msg = doc_to_chat_message(&doc);
            assert_eq!(msg.role, ChatRole::System);
        }
        #[test]
        fn doc_to_chat_message_defaults_to_user_for_unknown() {
            let mut doc = sample_message_doc().1;
            doc.insert("role", "SomethingElse");
            let msg = doc_to_chat_message(&doc);
            assert_eq!(msg.role, ChatRole::User);
        }
        #[test]
        fn doc_to_chat_message_extracts_content_and_timestamp() {
            let (_, doc) = sample_message_doc();
            let msg = doc_to_chat_message(&doc);
            assert_eq!(msg.content, "Hello there!");
            assert_eq!(msg.timestamp, "2025-01-01T12:00:00Z");
            assert_eq!(msg.session_id, "sess-1");
        }
        #[test]
        fn doc_to_chat_message_attachments_always_empty() {
            let (_, doc) = sample_message_doc();
            let msg = doc_to_chat_message(&doc);
            assert!(msg.attachments.is_empty());
        }
        // -- resolve_provider_url --
        const TEST_LITELLM_URL: &str = "http://localhost:4000";
        const TEST_LITELLM_MODEL: &str = "qwen3-32b";
        #[test]
        fn resolve_openai_returns_api_openai() {
            let (url, model) =
                resolve_provider_url(TEST_LITELLM_URL, TEST_LITELLM_MODEL, "openai", "gpt-4o");
            assert_eq!(url, "https://api.openai.com");
            assert_eq!(model, "gpt-4o");
        }
        #[test]
        fn resolve_anthropic_returns_api_anthropic() {
            let (url, model) = resolve_provider_url(
                TEST_LITELLM_URL,
                TEST_LITELLM_MODEL,
                "anthropic",
                "claude-3-opus",
            );
            assert_eq!(url, "https://api.anthropic.com");
            assert_eq!(model, "claude-3-opus");
        }
        #[test]
        fn resolve_huggingface_returns_model_url() {
            let (url, model) = resolve_provider_url(
                TEST_LITELLM_URL,
                TEST_LITELLM_MODEL,
                "huggingface",
                "meta-llama/Llama-2-7b",
            );
            assert_eq!(
                url,
                "https://api-inference.huggingface.co/models/meta-llama/Llama-2-7b"
            );
            assert_eq!(model, "meta-llama/Llama-2-7b");
        }
        #[test]
        fn resolve_unknown_defaults_to_litellm() {
            let (url, model) =
                resolve_provider_url(TEST_LITELLM_URL, TEST_LITELLM_MODEL, "litellm", "qwen3-32b");
            assert_eq!(url, TEST_LITELLM_URL);
            assert_eq!(model, "qwen3-32b");
        }
        #[test]
        fn resolve_empty_model_falls_back_to_server_default() {
            let (url, model) =
                resolve_provider_url(TEST_LITELLM_URL, TEST_LITELLM_MODEL, "litellm", "");
            assert_eq!(url, TEST_LITELLM_URL);
            assert_eq!(model, TEST_LITELLM_MODEL);
        }
    }
 }
--- a/src/infrastructure/config.rs
+++ b/src/infrastructure/config.rs
@@ -141,23 +141,19 @@ impl SmtpConfig {
 // ServiceUrls
 // ---------------------------------------------------------------------------
-/// URLs and credentials for external services (LiteLLM, SearXNG, S3, etc.).
+/// URLs and credentials for external services (Ollama, SearXNG, S3, etc.).
 #[derive(Debug)]
 pub struct ServiceUrls {
-    /// LiteLLM proxy base URL.
+    /// Ollama LLM instance base URL.
-    pub litellm_url: String,
+    pub ollama_url: String,
-    /// Default LiteLLM model to use.
+    /// Default Ollama model to use.
-    pub litellm_model: String,
+    pub ollama_model: String,
    /// LiteLLM API key for authenticated requests.
    pub litellm_api_key: String,
    /// SearXNG meta-search engine base URL.
    pub searxng_url: String,
    /// LangChain service URL.
    pub langchain_url: String,
    /// LangGraph service URL.
    pub langgraph_url: String,
    /// LangFlow visual workflow builder URL.
    pub langflow_url: String,
    /// Langfuse observability URL.
    pub langfuse_url: String,
    /// Vector database URL.
@@ -180,15 +176,13 @@ impl ServiceUrls {
    /// Currently infallible but returns `Result` for consistency.
    pub fn from_env() -> Result<Self, Error> {
        Ok(Self {
-            litellm_url: std::env::var("LITELLM_URL")
+            ollama_url: std::env::var("OLLAMA_URL")
-                .unwrap_or_else(|_| "http://localhost:4000".into()),
+                .unwrap_or_else(|_| "http://localhost:11434".into()),
-            litellm_model: std::env::var("LITELLM_MODEL").unwrap_or_else(|_| "qwen3-32b".into()),
+            ollama_model: std::env::var("OLLAMA_MODEL").unwrap_or_else(|_| "llama3.1:8b".into()),
            litellm_api_key: optional_env("LITELLM_API_KEY"),
            searxng_url: std::env::var("SEARXNG_URL")
                .unwrap_or_else(|_| "http://localhost:8888".into()),
            langchain_url: optional_env("LANGCHAIN_URL"),
            langgraph_url: optional_env("LANGGRAPH_URL"),
            langflow_url: optional_env("LANGFLOW_URL"),
            langfuse_url: optional_env("LANGFUSE_URL"),
            vectordb_url: optional_env("VECTORDB_URL"),
            s3_url: optional_env("S3_URL"),
@@ -234,7 +228,7 @@ impl StripeConfig {
 /// Comma-separated list of enabled LLM provider identifiers.
 ///
-/// For example: `LLM_PROVIDERS=litellm,openai,anthropic`
+/// For example: `LLM_PROVIDERS=ollama,openai,anthropic`
 #[derive(Debug)]
 pub struct LlmProvidersConfig {
    /// Parsed provider names.
@@ -257,160 +251,3 @@ impl LlmProvidersConfig {
        Ok(Self { providers })
    }
 }
 #[cfg(test)]
 mod tests {
    #![allow(clippy::unwrap_used, clippy::expect_used)]
    use super::*;
    use pretty_assertions::assert_eq;
    use serial_test::serial;
    // -----------------------------------------------------------------------
    // KeycloakConfig endpoint methods (no env vars needed)
    // -----------------------------------------------------------------------
    fn sample_keycloak() -> KeycloakConfig {
        KeycloakConfig {
            url: "https://auth.example.com".into(),
            realm: "myrealm".into(),
            client_id: "dashboard".into(),
            redirect_uri: "https://app.example.com/callback".into(),
            app_url: "https://app.example.com".into(),
            admin_client_id: String::new(),
            admin_client_secret: SecretString::from(String::new()),
        }
    }
    #[test]
    fn keycloak_auth_endpoint() {
        let kc = sample_keycloak();
        assert_eq!(
            kc.auth_endpoint(),
            "https://auth.example.com/realms/myrealm/protocol/openid-connect/auth"
        );
    }
    #[test]
    fn keycloak_token_endpoint() {
        let kc = sample_keycloak();
        assert_eq!(
            kc.token_endpoint(),
            "https://auth.example.com/realms/myrealm/protocol/openid-connect/token"
        );
    }
    #[test]
    fn keycloak_userinfo_endpoint() {
        let kc = sample_keycloak();
        assert_eq!(
            kc.userinfo_endpoint(),
            "https://auth.example.com/realms/myrealm/protocol/openid-connect/userinfo"
        );
    }
    #[test]
    fn keycloak_logout_endpoint() {
        let kc = sample_keycloak();
        assert_eq!(
            kc.logout_endpoint(),
            "https://auth.example.com/realms/myrealm/protocol/openid-connect/logout"
        );
    }
    // -----------------------------------------------------------------------
    // LlmProvidersConfig::from_env()
    // -----------------------------------------------------------------------
    #[test]
    #[serial]
    fn llm_providers_empty_string() {
        std::env::set_var("LLM_PROVIDERS", "");
        let cfg = LlmProvidersConfig::from_env().unwrap();
        assert!(cfg.providers.is_empty());
        std::env::remove_var("LLM_PROVIDERS");
    }
    #[test]
    #[serial]
    fn llm_providers_single() {
        std::env::set_var("LLM_PROVIDERS", "litellm");
        let cfg = LlmProvidersConfig::from_env().unwrap();
        assert_eq!(cfg.providers, vec!["litellm"]);
        std::env::remove_var("LLM_PROVIDERS");
    }
    #[test]
    #[serial]
    fn llm_providers_multiple() {
        std::env::set_var("LLM_PROVIDERS", "litellm,openai,anthropic");
        let cfg = LlmProvidersConfig::from_env().unwrap();
        assert_eq!(cfg.providers, vec!["litellm", "openai", "anthropic"]);
        std::env::remove_var("LLM_PROVIDERS");
    }
    #[test]
    #[serial]
    fn llm_providers_trims_whitespace() {
        std::env::set_var("LLM_PROVIDERS", " litellm , openai ");
        let cfg = LlmProvidersConfig::from_env().unwrap();
        assert_eq!(cfg.providers, vec!["litellm", "openai"]);
        std::env::remove_var("LLM_PROVIDERS");
    }
    #[test]
    #[serial]
    fn llm_providers_filters_empty_entries() {
        std::env::set_var("LLM_PROVIDERS", "litellm,,openai,");
        let cfg = LlmProvidersConfig::from_env().unwrap();
        assert_eq!(cfg.providers, vec!["litellm", "openai"]);
        std::env::remove_var("LLM_PROVIDERS");
    }
    // -----------------------------------------------------------------------
    // ServiceUrls::from_env() defaults
    // -----------------------------------------------------------------------
    #[test]
    #[serial]
    fn service_urls_default_litellm_url() {
        std::env::remove_var("LITELLM_URL");
        let svc = ServiceUrls::from_env().unwrap();
        assert_eq!(svc.litellm_url, "http://localhost:4000");
    }
    #[test]
    #[serial]
    fn service_urls_default_litellm_model() {
        std::env::remove_var("LITELLM_MODEL");
        let svc = ServiceUrls::from_env().unwrap();
        assert_eq!(svc.litellm_model, "qwen3-32b");
    }
    #[test]
    #[serial]
    fn service_urls_default_searxng_url() {
        std::env::remove_var("SEARXNG_URL");
        let svc = ServiceUrls::from_env().unwrap();
        assert_eq!(svc.searxng_url, "http://localhost:8888");
    }
    #[test]
    #[serial]
    fn service_urls_custom_litellm_url() {
        std::env::set_var("LITELLM_URL", "http://litellm-host:4000");
        let svc = ServiceUrls::from_env().unwrap();
        assert_eq!(svc.litellm_url, "http://litellm-host:4000");
        std::env::remove_var("LITELLM_URL");
    }
    #[test]
    #[serial]
    fn required_env_missing_returns_config_error() {
        std::env::remove_var("__TEST_REQUIRED_MISSING__");
        let result = required_env("__TEST_REQUIRED_MISSING__");
        assert!(result.is_err());
        let err_msg = result.unwrap_err().to_string();
        assert!(err_msg.contains("__TEST_REQUIRED_MISSING__"));
    }
 }
--- a/src/infrastructure/error.rs
+++ b/src/infrastructure/error.rs
@@ -41,53 +41,3 @@ impl IntoResponse for Error {
        }
    }
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use axum::response::IntoResponse;
    use pretty_assertions::assert_eq;
    #[test]
    fn state_error_display() {
        let err = Error::StateError("bad state".into());
        assert_eq!(err.to_string(), "bad state");
    }
    #[test]
    fn database_error_display() {
        let err = Error::DatabaseError("connection lost".into());
        assert_eq!(err.to_string(), "database error: connection lost");
    }
    #[test]
    fn config_error_display() {
        let err = Error::ConfigError("missing var".into());
        assert_eq!(err.to_string(), "configuration error: missing var");
    }
    #[test]
    fn state_error_into_response_500() {
        let resp = Error::StateError("oops".into()).into_response();
        assert_eq!(resp.status(), StatusCode::INTERNAL_SERVER_ERROR);
    }
    #[test]
    fn database_error_into_response_503() {
        let resp = Error::DatabaseError("down".into()).into_response();
        assert_eq!(resp.status(), StatusCode::SERVICE_UNAVAILABLE);
    }
    #[test]
    fn config_error_into_response_500() {
        let resp = Error::ConfigError("bad cfg".into()).into_response();
        assert_eq!(resp.status(), StatusCode::INTERNAL_SERVER_ERROR);
    }
    #[test]
    fn io_error_into_response_500() {
        let io_err = std::io::Error::new(std::io::ErrorKind::NotFound, "not found");
        let resp = Error::IoError(io_err).into_response();
        assert_eq!(resp.status(), StatusCode::INTERNAL_SERVER_ERROR);
    }
 }
--- a/src/infrastructure/langgraph.rs
+++ b/src/infrastructure/langgraph.rs
@@ -1,108 +0,0 @@
 use dioxus::prelude::*;
 #[cfg(feature = "server")]
 use serde::Deserialize;
 use crate::models::AgentEntry;
 /// Raw assistant object returned by the LangGraph `POST /assistants/search`
 /// endpoint. Only the fields we display are deserialized; unknown keys are
 /// silently ignored thanks to serde defaults.
 #[cfg(feature = "server")]
 #[derive(Deserialize)]
 struct LangGraphAssistant {
    assistant_id: String,
    #[serde(default)]
    name: String,
    #[serde(default)]
    graph_id: String,
    #[serde(default)]
    metadata: serde_json::Value,
 }
 /// Fetch the list of assistants (agents) from a LangGraph instance.
 ///
 /// Calls `POST <langgraph_url>/assistants/search` with an empty body to
 /// retrieve every registered assistant. Each result is mapped to the
 /// frontend-friendly `AgentEntry` model.
 ///
 /// # Returns
 ///
 /// A vector of `AgentEntry` structs. Returns an empty vector when the
 /// LangGraph URL is not configured or the service is unreachable.
 ///
 /// # Errors
 ///
 /// Returns `ServerFnError` on network or deserialization failures that
 /// indicate a misconfigured (but present) LangGraph instance.
 #[server(endpoint = "list-langgraph-agents")]
 pub async fn list_langgraph_agents() -> Result<Vec<AgentEntry>, ServerFnError> {
    let state: crate::infrastructure::ServerState =
        dioxus_fullstack::FullstackContext::extract().await?;
    let base_url = state.services.langgraph_url.clone();
    if base_url.is_empty() {
        return Ok(Vec::new());
    }
    let url = format!("{}/assistants/search", base_url.trim_end_matches('/'));
    let client = reqwest::Client::builder()
        .timeout(std::time::Duration::from_secs(5))
        .build()
        .map_err(|e| ServerFnError::new(format!("HTTP client error: {e}")))?;
    // LangGraph expects a POST with a JSON body (empty object = no filters).
    let resp = match client
        .post(&url)
        .header("content-type", "application/json")
        .body("{}")
        .send()
        .await
    {
        Ok(r) if r.status().is_success() => r,
        Ok(r) => {
            let status = r.status();
            let body = r.text().await.unwrap_or_default();
            tracing::error!("LangGraph returned {status}: {body}");
            return Ok(Vec::new());
        }
        Err(e) => {
            tracing::error!("LangGraph request failed: {e}");
            return Ok(Vec::new());
        }
    };
    let assistants: Vec<LangGraphAssistant> = resp
        .json()
        .await
        .map_err(|e| ServerFnError::new(format!("Failed to parse LangGraph response: {e}")))?;
    let entries = assistants
        .into_iter()
        .map(|a| {
            // Use the assistant name if present, otherwise fall back to graph_id.
            let name = if a.name.is_empty() {
                a.graph_id.clone()
            } else {
                a.name
            };
            // Extract a description from metadata if available.
            let description = a
                .metadata
                .get("description")
                .and_then(|v| v.as_str())
                .unwrap_or("")
                .to_string();
            AgentEntry {
                id: a.assistant_id,
                name,
                description,
                status: "active".to_string(),
            }
        })
        .collect();
    Ok(entries)
 }
--- a/src/infrastructure/litellm.rs
+++ b/src/infrastructure/litellm.rs
@@ -1,403 +0,0 @@
 #[cfg(feature = "server")]
 use std::collections::HashMap;
 use dioxus::prelude::*;
 use serde::{Deserialize, Serialize};
 use crate::models::LitellmUsageStats;
 #[cfg(feature = "server")]
 use crate::models::ModelUsage;
 /// Status of a LiteLLM proxy instance, including connectivity and available models.
 ///
 /// # Fields
 ///
 /// * `online` - Whether the LiteLLM API responded successfully
 /// * `models` - List of model IDs available through the proxy
 #[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
 pub struct LitellmStatus {
    pub online: bool,
    pub models: Vec<String>,
 }
 /// Response from LiteLLM's `GET /v1/models` endpoint (OpenAI-compatible).
 #[cfg(feature = "server")]
 #[derive(Deserialize)]
 struct ModelsResponse {
    data: Vec<ModelObject>,
 }
 /// A single model entry from the OpenAI-compatible models list.
 #[cfg(feature = "server")]
 #[derive(Deserialize)]
 struct ModelObject {
    id: String,
 }
 /// Check the status of a LiteLLM proxy by querying its models endpoint.
 ///
 /// Calls `GET <litellm_url>/v1/models` to list available models and determine
 /// whether the instance is reachable. Sends the API key as a Bearer token
 /// if configured.
 ///
 /// # Arguments
 ///
 /// * `litellm_url` - Base URL of the LiteLLM proxy (e.g. "http://localhost:4000")
 ///
 /// # Returns
 ///
 /// A `LitellmStatus` with `online: true` and model IDs if reachable,
 /// or `online: false` with an empty model list on failure
 ///
 /// # Errors
 ///
 /// Returns `ServerFnError` only on serialization issues; network failures
 /// are caught and returned as `online: false`
 #[post("/api/litellm-status")]
 pub async fn get_litellm_status(litellm_url: String) -> Result<LitellmStatus, ServerFnError> {
    let state: crate::infrastructure::ServerState =
        dioxus_fullstack::FullstackContext::extract().await?;
    let base_url = if litellm_url.is_empty() {
        state.services.litellm_url.clone()
    } else {
        litellm_url
    };
    let api_key = state.services.litellm_api_key.clone();
    let url = format!("{}/v1/models", base_url.trim_end_matches('/'));
    let client = reqwest::Client::builder()
        .timeout(std::time::Duration::from_secs(5))
        .build()
        .map_err(|e| ServerFnError::new(format!("HTTP client error: {e}")))?;
    let mut request = client.get(&url);
    if !api_key.is_empty() {
        request = request.header("Authorization", format!("Bearer {api_key}"));
    }
    let resp = match request.send().await {
        Ok(r) if r.status().is_success() => r,
        _ => {
            return Ok(LitellmStatus {
                online: false,
                models: Vec::new(),
            });
        }
    };
    let body: ModelsResponse = match resp.json().await {
        Ok(b) => b,
        Err(_) => {
            return Ok(LitellmStatus {
                online: true,
                models: Vec::new(),
            });
        }
    };
    let models = body.data.into_iter().map(|m| m.id).collect();
    Ok(LitellmStatus {
        online: true,
        models,
    })
 }
 /// Response from LiteLLM's `GET /global/activity` endpoint.
 ///
 /// Returns aggregate token counts and API request totals for a date range.
 /// Available on the free tier (no Enterprise license needed).
 #[cfg(feature = "server")]
 #[derive(Debug, Deserialize)]
 struct ActivityResponse {
    /// Total tokens across all models in the date range
    #[serde(default)]
    sum_total_tokens: u64,
 }
 /// Per-model entry from `GET /global/activity/model`.
 ///
 /// Each entry contains a model name and its aggregated token total.
 #[cfg(feature = "server")]
 #[derive(Debug, Deserialize)]
 struct ActivityModelEntry {
    /// Model identifier (may be empty for unattributed traffic)
    #[serde(default)]
    model: String,
    /// Sum of tokens used by this model in the date range
    #[serde(default)]
    sum_total_tokens: u64,
 }
 /// Per-model spend entry from `GET /global/spend/models`.
 ///
 /// Each entry maps a model name to its total spend in USD.
 #[cfg(feature = "server")]
 #[derive(Debug, Deserialize)]
 struct SpendModelEntry {
    /// Model identifier
    #[serde(default)]
    model: String,
    /// Total spend in USD
    #[serde(default)]
    total_spend: f64,
 }
 /// Merge per-model token counts and spend data into `ModelUsage` entries.
 ///
 /// Joins `activity_models` (tokens) and `spend_models` (spend) by model
 /// name using a HashMap for O(n + m) merge. Entries with empty model
 /// names are skipped.
 ///
 /// # Arguments
 ///
 /// * `activity_models` - Per-model token data from `/global/activity/model`
 /// * `spend_models` - Per-model spend data from `/global/spend/models`
 ///
 /// # Returns
 ///
 /// Merged list sorted by total tokens descending
 #[cfg(feature = "server")]
 fn merge_model_data(
    activity_models: Vec<ActivityModelEntry>,
    spend_models: Vec<SpendModelEntry>,
 ) -> Vec<ModelUsage> {
    let mut model_map: HashMap<String, ModelUsage> = HashMap::new();
    for entry in activity_models {
        if entry.model.is_empty() {
            continue;
        }
        model_map
            .entry(entry.model.clone())
            .or_insert_with(|| ModelUsage {
                model: entry.model,
                ..Default::default()
            })
            .total_tokens = entry.sum_total_tokens;
    }
    for entry in spend_models {
        if entry.model.is_empty() {
            continue;
        }
        model_map
            .entry(entry.model.clone())
            .or_insert_with(|| ModelUsage {
                model: entry.model,
                ..Default::default()
            })
            .spend = entry.total_spend;
    }
    let mut result: Vec<ModelUsage> = model_map.into_values().collect();
    result.sort_by(|a, b| b.total_tokens.cmp(&a.total_tokens));
    result
 }
 /// Fetch aggregated usage statistics from LiteLLM's free-tier APIs.
 ///
 /// Combines three endpoints to build a complete usage picture:
 /// - `GET /global/activity` - total token counts
 /// - `GET /global/activity/model` - per-model token breakdown
 /// - `GET /global/spend/models` - per-model spend in USD
 ///
 /// # Arguments
 ///
 /// * `start_date` - Start of the reporting period in `YYYY-MM-DD` format
 /// * `end_date` - End of the reporting period in `YYYY-MM-DD` format
 ///
 /// # Returns
 ///
 /// Aggregated usage stats; returns default (zeroed) stats on network
 /// failure or permission errors
 ///
 /// # Errors
 ///
 /// Returns `ServerFnError` only on HTTP client construction failure
 #[post("/api/litellm-usage")]
 pub async fn get_litellm_usage(
    start_date: String,
    end_date: String,
 ) -> Result<LitellmUsageStats, ServerFnError> {
    let state: crate::infrastructure::ServerState =
        dioxus_fullstack::FullstackContext::extract().await?;
    let base_url = &state.services.litellm_url;
    let api_key = &state.services.litellm_api_key;
    if base_url.is_empty() {
        return Ok(LitellmUsageStats::default());
    }
    let base = base_url.trim_end_matches('/');
    let date_params = format!("start_date={start_date}&end_date={end_date}");
    let client = reqwest::Client::builder()
        .timeout(std::time::Duration::from_secs(10))
        .build()
        .map_err(|e| ServerFnError::new(format!("HTTP client error: {e}")))?;
    // Helper closure to build an authenticated GET request
    let auth_get = |url: String| {
        let mut req = client.get(url);
        if !api_key.is_empty() {
            req = req.header("Authorization", format!("Bearer {api_key}"));
        }
        req
    };
    // Fire all three requests concurrently to minimise latency
    let (activity_res, model_activity_res, model_spend_res) = tokio::join!(
        auth_get(format!("{base}/global/activity?{date_params}")).send(),
        auth_get(format!("{base}/global/activity/model?{date_params}")).send(),
        auth_get(format!("{base}/global/spend/models?{date_params}")).send(),
    );
    // Parse total token count from /global/activity
    let total_tokens = match activity_res {
        Ok(r) if r.status().is_success() => r
            .json::<ActivityResponse>()
            .await
            .map(|a| a.sum_total_tokens)
            .unwrap_or(0),
        _ => 0,
    };
    // Parse per-model token breakdown from /global/activity/model
    let activity_models: Vec<ActivityModelEntry> = match model_activity_res {
        Ok(r) if r.status().is_success() => r.json().await.unwrap_or_default(),
        _ => Vec::new(),
    };
    // Parse per-model spend from /global/spend/models
    let spend_models: Vec<SpendModelEntry> = match model_spend_res {
        Ok(r) if r.status().is_success() => r.json().await.unwrap_or_default(),
        _ => Vec::new(),
    };
    let total_spend: f64 = spend_models.iter().map(|m| m.total_spend).sum();
    let model_breakdown = merge_model_data(activity_models, spend_models);
    Ok(LitellmUsageStats {
        total_spend,
        // Free-tier endpoints don't provide prompt/completion split;
        // total_tokens comes from /global/activity.
        total_prompt_tokens: 0,
        total_completion_tokens: 0,
        total_tokens,
        model_breakdown,
    })
 }
 #[cfg(all(test, feature = "server"))]
 mod tests {
    use super::*;
    #[test]
    fn merge_empty_inputs() {
        let result = merge_model_data(Vec::new(), Vec::new());
        assert!(result.is_empty());
    }
    #[test]
    fn merge_activity_only() {
        let activity = vec![ActivityModelEntry {
            model: "gpt-4".into(),
            sum_total_tokens: 1500,
        }];
        let result = merge_model_data(activity, Vec::new());
        assert_eq!(result.len(), 1);
        assert_eq!(result[0].model, "gpt-4");
        assert_eq!(result[0].total_tokens, 1500);
        assert_eq!(result[0].spend, 0.0);
    }
    #[test]
    fn merge_spend_only() {
        let spend = vec![SpendModelEntry {
            model: "gpt-4".into(),
            total_spend: 2.5,
        }];
        let result = merge_model_data(Vec::new(), spend);
        assert_eq!(result.len(), 1);
        assert_eq!(result[0].model, "gpt-4");
        assert_eq!(result[0].spend, 2.5);
        assert_eq!(result[0].total_tokens, 0);
    }
    #[test]
    fn merge_joins_by_model_name() {
        let activity = vec![
            ActivityModelEntry {
                model: "gpt-4".into(),
                sum_total_tokens: 5000,
            },
            ActivityModelEntry {
                model: "claude-3".into(),
                sum_total_tokens: 3000,
            },
        ];
        let spend = vec![
            SpendModelEntry {
                model: "gpt-4".into(),
                total_spend: 1.0,
            },
            SpendModelEntry {
                model: "claude-3".into(),
                total_spend: 0.5,
            },
        ];
        let result = merge_model_data(activity, spend);
        assert_eq!(result.len(), 2);
        // Sorted by tokens descending: gpt-4 (5000) before claude-3 (3000)
        assert_eq!(result[0].model, "gpt-4");
        assert_eq!(result[0].total_tokens, 5000);
        assert_eq!(result[0].spend, 1.0);
        assert_eq!(result[1].model, "claude-3");
        assert_eq!(result[1].total_tokens, 3000);
        assert_eq!(result[1].spend, 0.5);
    }
    #[test]
    fn merge_skips_empty_model_names() {
        let activity = vec![
            ActivityModelEntry {
                model: "".into(),
                sum_total_tokens: 100,
            },
            ActivityModelEntry {
                model: "gpt-4".into(),
                sum_total_tokens: 500,
            },
        ];
        let spend = vec![SpendModelEntry {
            model: "".into(),
            total_spend: 0.01,
        }];
        let result = merge_model_data(activity, spend);
        assert_eq!(result.len(), 1);
        assert_eq!(result[0].model, "gpt-4");
    }
    #[test]
    fn merge_unmatched_models_appear_in_both_directions() {
        let activity = vec![ActivityModelEntry {
            model: "tokens-only".into(),
            sum_total_tokens: 1000,
        }];
        let spend = vec![SpendModelEntry {
            model: "spend-only".into(),
            total_spend: 0.5,
        }];
        let result = merge_model_data(activity, spend);
        assert_eq!(result.len(), 2);
        // tokens-only has 1000 tokens, spend-only has 0 tokens
        assert_eq!(result[0].model, "tokens-only");
        assert_eq!(result[0].total_tokens, 1000);
        assert_eq!(result[1].model, "spend-only");
        assert_eq!(result[1].spend, 0.5);
    }
 }
--- a/src/infrastructure/llm.rs
+++ b/src/infrastructure/llm.rs
@@ -4,23 +4,23 @@ use dioxus::prelude::*;
 mod inner {
    use serde::{Deserialize, Serialize};
-    /// A single message in the OpenAI-compatible chat format used by LiteLLM.
+    /// A single message in the OpenAI-compatible chat format used by Ollama.
    #[derive(Serialize)]
    pub(super) struct ChatMessage {
        pub role: String,
        pub content: String,
    }
-    /// Request body for the OpenAI-compatible chat completions endpoint.
+    /// Request body for Ollama's OpenAI-compatible chat completions endpoint.
    #[derive(Serialize)]
-    pub(super) struct ChatCompletionRequest {
+    pub(super) struct OllamaChatRequest {
        pub model: String,
        pub messages: Vec<ChatMessage>,
        /// Disable streaming so we get a single JSON response.
        pub stream: bool,
    }
-    /// A single choice in the chat completions response.
+    /// A single choice in the Ollama chat completions response.
    #[derive(Deserialize)]
    pub(super) struct ChatChoice {
        pub message: ChatResponseMessage,
@@ -32,9 +32,9 @@ mod inner {
        pub content: String,
    }
-    /// Top-level response from the `/v1/chat/completions` endpoint.
+    /// Top-level response from Ollama's `/v1/chat/completions` endpoint.
    #[derive(Deserialize)]
-    pub(super) struct ChatCompletionResponse {
+    pub(super) struct OllamaChatResponse {
        pub choices: Vec<ChatChoice>,
    }
@@ -72,25 +72,7 @@ mod inner {
        }
        let html = resp.text().await.ok()?;
-        parse_article_html(&html)
+        let document = scraper::Html::parse_document(&html);
    }
    /// Parse article text from raw HTML without any network I/O.
    ///
    /// Uses a tiered extraction strategy:
    /// 1. Try content within `<article>`, `<main>`, or `[role="main"]`
    /// 2. Fall back to all `<p>` tags outside excluded containers
    ///
    /// # Arguments
    ///
    /// * `html` - Raw HTML string to parse
    ///
    /// # Returns
    ///
    /// The extracted text, or `None` if extraction yields < 100 chars.
    /// Output is capped at 8000 characters.
    pub(crate) fn parse_article_html(html: &str) -> Option<String> {
        let document = scraper::Html::parse_document(html);
        // Strategy 1: Extract from semantic article containers.
        // Most news sites wrap the main content in <article>, <main>,
@@ -152,12 +134,12 @@ mod inner {
    }
    /// Sum the total character length of all collected text parts.
-    pub(crate) fn joined_len(parts: &[String]) -> usize {
+    fn joined_len(parts: &[String]) -> usize {
        parts.iter().map(|s| s.len()).sum()
    }
 }
-/// Summarize an article using a LiteLLM proxy.
+/// Summarize an article using a local Ollama instance.
 ///
 /// First attempts to fetch the full article text from the provided URL.
 /// If that fails (paywall, timeout, etc.), falls back to the search snippet.
@@ -167,8 +149,8 @@ mod inner {
 ///
 /// * `snippet` - The search result snippet (fallback content)
 /// * `article_url` - The original article URL to fetch full text from
-/// * `litellm_url` - Base URL of the LiteLLM proxy (e.g. "http://localhost:4000")
+/// * `ollama_url` - Base URL of the Ollama instance (e.g. "http://localhost:11434")
-/// * `model` - The model ID to use (e.g. "qwen3-32b")
+/// * `model` - The Ollama model ID to use (e.g. "llama3.1:8b")
 ///
 /// # Returns
 ///
@@ -176,38 +158,36 @@ mod inner {
 ///
 /// # Errors
 ///
-/// Returns `ServerFnError` if the LiteLLM request fails or response parsing fails
+/// Returns `ServerFnError` if the Ollama request fails or response parsing fails
 #[post("/api/summarize")]
 pub async fn summarize_article(
    snippet: String,
    article_url: String,
-    litellm_url: String,
+    ollama_url: String,
    model: String,
 ) -> Result<String, ServerFnError> {
-    use inner::{fetch_article_text, ChatCompletionRequest, ChatCompletionResponse, ChatMessage};
+    use inner::{fetch_article_text, ChatMessage, OllamaChatRequest, OllamaChatResponse};
    let state: crate::infrastructure::ServerState =
        dioxus_fullstack::FullstackContext::extract().await?;
    // Use caller-provided values or fall back to ServerState config
-    let base_url = if litellm_url.is_empty() {
+    let base_url = if ollama_url.is_empty() {
-        state.services.litellm_url.clone()
+        state.services.ollama_url.clone()
    } else {
-        litellm_url
+        ollama_url
    };
    let model = if model.is_empty() {
-        state.services.litellm_model.clone()
+        state.services.ollama_model.clone()
    } else {
        model
    };
    let api_key = state.services.litellm_api_key.clone();
    // Try to fetch the full article; fall back to the search snippet
    let article_text = fetch_article_text(&article_url).await.unwrap_or(snippet);
-    let request_body = ChatCompletionRequest {
+    let request_body = OllamaChatRequest {
        model,
        stream: false,
        messages: vec![ChatMessage {
@@ -225,48 +205,42 @@ pub async fn summarize_article(
    let url = format!("{}/v1/chat/completions", base_url.trim_end_matches('/'));
    let client = reqwest::Client::new();
-    let mut request = client
+    let resp = client
        .post(&url)
        .header("content-type", "application/json")
-        .json(&request_body);
+        .json(&request_body)
    if !api_key.is_empty() {
        request = request.header("Authorization", format!("Bearer {api_key}"));
    }
    let resp = request
        .send()
        .await
-        .map_err(|e| ServerFnError::new(format!("LiteLLM request failed: {e}")))?;
+        .map_err(|e| ServerFnError::new(format!("Ollama request failed: {e}")))?;
    if !resp.status().is_success() {
        let status = resp.status();
        let body = resp.text().await.unwrap_or_default();
        return Err(ServerFnError::new(format!(
-            "LiteLLM returned {status}: {body}"
+            "Ollama returned {status}: {body}"
        )));
    }
-    let body: ChatCompletionResponse = resp
+    let body: OllamaChatResponse = resp
        .json()
        .await
-        .map_err(|e| ServerFnError::new(format!("Failed to parse LiteLLM response: {e}")))?;
+        .map_err(|e| ServerFnError::new(format!("Failed to parse Ollama response: {e}")))?;
    body.choices
        .first()
        .map(|choice| choice.message.content.clone())
-        .ok_or_else(|| ServerFnError::new("Empty response from LiteLLM"))
+        .ok_or_else(|| ServerFnError::new("Empty response from Ollama"))
 }
 /// A lightweight chat message for the follow-up conversation.
-/// Uses simple String role ("system"/"user"/"assistant") for OpenAI compatibility.
+/// Uses simple String role ("system"/"user"/"assistant") for Ollama compatibility.
 #[derive(Debug, Clone, PartialEq, serde::Serialize, serde::Deserialize)]
 pub struct FollowUpMessage {
    pub role: String,
    pub content: String,
 }
-/// Send a follow-up question about an article using a LiteLLM proxy.
+/// Send a follow-up question about an article using a local Ollama instance.
 ///
 /// Accepts the full conversation history (system context + prior turns) and
 /// returns the assistant's next response. The system message should contain
@@ -275,8 +249,8 @@ pub struct FollowUpMessage {
 /// # Arguments
 ///
 /// * `messages` - The conversation history including system context
-/// * `litellm_url` - Base URL of the LiteLLM proxy
+/// * `ollama_url` - Base URL of the Ollama instance
-/// * `model` - The model ID to use
+/// * `model` - The Ollama model ID to use
 ///
 /// # Returns
 ///
@@ -284,32 +258,30 @@ pub struct FollowUpMessage {
 ///
 /// # Errors
 ///
-/// Returns `ServerFnError` if the LiteLLM request fails or response parsing fails
+/// Returns `ServerFnError` if the Ollama request fails or response parsing fails
 #[post("/api/chat")]
 pub async fn chat_followup(
    messages: Vec<FollowUpMessage>,
-    litellm_url: String,
+    ollama_url: String,
    model: String,
 ) -> Result<String, ServerFnError> {
-    use inner::{ChatCompletionRequest, ChatCompletionResponse, ChatMessage};
+    use inner::{ChatMessage, OllamaChatRequest, OllamaChatResponse};
    let state: crate::infrastructure::ServerState =
        dioxus_fullstack::FullstackContext::extract().await?;
-    let base_url = if litellm_url.is_empty() {
+    let base_url = if ollama_url.is_empty() {
-        state.services.litellm_url.clone()
+        state.services.ollama_url.clone()
    } else {
-        litellm_url
+        ollama_url
    };
    let model = if model.is_empty() {
-        state.services.litellm_model.clone()
+        state.services.ollama_model.clone()
    } else {
        model
    };
    let api_key = state.services.litellm_api_key.clone();
    // Convert FollowUpMessage to inner ChatMessage for the request
    let chat_messages: Vec<ChatMessage> = messages
        .into_iter()
@@ -319,7 +291,7 @@ pub async fn chat_followup(
        })
        .collect();
-    let request_body = ChatCompletionRequest {
+    let request_body = OllamaChatRequest {
        model,
        stream: false,
        messages: chat_messages,
@@ -327,182 +299,29 @@ pub async fn chat_followup(
    let url = format!("{}/v1/chat/completions", base_url.trim_end_matches('/'));
    let client = reqwest::Client::new();
-    let mut request = client
+    let resp = client
        .post(&url)
        .header("content-type", "application/json")
-        .json(&request_body);
+        .json(&request_body)
    if !api_key.is_empty() {
        request = request.header("Authorization", format!("Bearer {api_key}"));
    }
    let resp = request
        .send()
        .await
-        .map_err(|e| ServerFnError::new(format!("LiteLLM request failed: {e}")))?;
+        .map_err(|e| ServerFnError::new(format!("Ollama request failed: {e}")))?;
    if !resp.status().is_success() {
        let status = resp.status();
        let body = resp.text().await.unwrap_or_default();
        return Err(ServerFnError::new(format!(
-            "LiteLLM returned {status}: {body}"
+            "Ollama returned {status}: {body}"
        )));
    }
-    let body: ChatCompletionResponse = resp
+    let body: OllamaChatResponse = resp
        .json()
        .await
-        .map_err(|e| ServerFnError::new(format!("Failed to parse LiteLLM response: {e}")))?;
+        .map_err(|e| ServerFnError::new(format!("Failed to parse Ollama response: {e}")))?;
    body.choices
        .first()
        .map(|choice| choice.message.content.clone())
-        .ok_or_else(|| ServerFnError::new("Empty response from LiteLLM"))
+        .ok_or_else(|| ServerFnError::new("Empty response from Ollama"))
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use pretty_assertions::assert_eq;
    // -----------------------------------------------------------------------
    // FollowUpMessage serde tests
    // -----------------------------------------------------------------------
    #[test]
    fn followup_message_serde_round_trip() {
        let msg = FollowUpMessage {
            role: "assistant".into(),
            content: "Here is my answer.".into(),
        };
        let json = serde_json::to_string(&msg).expect("serialize FollowUpMessage");
        let back: FollowUpMessage =
            serde_json::from_str(&json).expect("deserialize FollowUpMessage");
        assert_eq!(msg, back);
    }
    #[test]
    fn followup_message_deserialize_from_json_literal() {
        let json = r#"{"role":"system","content":"You are helpful."}"#;
        let msg: FollowUpMessage = serde_json::from_str(json).expect("deserialize literal");
        assert_eq!(msg.role, "system");
        assert_eq!(msg.content, "You are helpful.");
    }
    // -----------------------------------------------------------------------
    // joined_len and parse_article_html tests (server feature required)
    // -----------------------------------------------------------------------
    #[cfg(feature = "server")]
    mod server_tests {
        use super::super::inner::{joined_len, parse_article_html};
        use pretty_assertions::assert_eq;
        #[test]
        fn joined_len_empty_input() {
            assert_eq!(joined_len(&[]), 0);
        }
        #[test]
        fn joined_len_sums_correctly() {
            let parts = vec!["abc".into(), "de".into(), "fghij".into()];
            assert_eq!(joined_len(&parts), 10);
        }
        // -------------------------------------------------------------------
        // parse_article_html tests
        // -------------------------------------------------------------------
        // Helper: generate a string of given length from a repeated word.
        fn lorem(len: usize) -> String {
            "Lorem ipsum dolor sit amet consectetur adipiscing elit "
                .repeat((len / 55) + 1)
                .chars()
                .take(len)
                .collect()
        }
        #[test]
        fn article_tag_extracts_text() {
            let body = lorem(250);
            let html = format!("<html><body><article><p>{body}</p></article></body></html>");
            let result = parse_article_html(&html);
            assert!(result.is_some(), "expected Some for article tag");
            assert!(result.unwrap().contains("Lorem"));
        }
        #[test]
        fn main_tag_extracts_text() {
            let body = lorem(250);
            let html = format!("<html><body><main><p>{body}</p></main></body></html>");
            let result = parse_article_html(&html);
            assert!(result.is_some(), "expected Some for main tag");
        }
        #[test]
        fn fallback_to_p_tags_when_article_main_yield_little() {
            // No <article>/<main>, so falls back to <p> tags
            let body = lorem(250);
            let html = format!("<html><body><div><p>{body}</p></div></body></html>");
            let result = parse_article_html(&html);
            assert!(result.is_some(), "expected fallback to <p> tags");
        }
        #[test]
        fn excludes_nav_footer_aside_content() {
            // Content only inside excluded containers -- should be excluded
            let body = lorem(250);
            let html = format!(
                "<html><body>\
                 <nav><p>{body}</p></nav>\
                 <footer><p>{body}</p></footer>\
                 <aside><p>{body}</p></aside>\
                 </body></html>"
            );
            let result = parse_article_html(&html);
            assert!(result.is_none(), "expected None for excluded-only content");
        }
        #[test]
        fn returns_none_when_text_too_short() {
            let html = "<html><body><p>Short.</p></body></html>";
            let result = parse_article_html(html);
            assert!(result.is_none(), "expected None for short text");
        }
        #[test]
        fn truncates_at_8000_chars() {
            let body = lorem(10000);
            let html = format!("<html><body><article><p>{body}</p></article></body></html>");
            let result = parse_article_html(&html).expect("expected Some");
            assert!(
                result.len() <= 8000,
                "expected <= 8000 chars, got {}",
                result.len()
            );
        }
        #[test]
        fn skips_fragments_under_30_chars() {
            // Only fragments < 30 chars -- should yield None
            let html = "<html><body><article>\
                         <p>Short frag one</p>\
                         <p>Another tiny bit</p>\
                         </article></body></html>";
            let result = parse_article_html(html);
            assert!(result.is_none(), "expected None for tiny fragments");
        }
        #[test]
        fn extracts_from_role_main_attribute() {
            let body = lorem(250);
            let html = format!(
                "<html><body>\
                 <div role=\"main\"><p>{body}</p></div>\
                 </body></html>"
            );
            let result = parse_article_html(&html);
            assert!(result.is_some(), "expected Some for role=main");
        }
    }
 }
--- a/src/infrastructure/mod.rs
+++ b/src/infrastructure/mod.rs
@@ -2,9 +2,8 @@
 // the #[server] macro generates client stubs for the web target)
 pub mod auth_check;
 pub mod chat;
 pub mod langgraph;
 pub mod litellm;
 pub mod llm;
 pub mod ollama;
 pub mod searxng;
 // Server-only modules (Axum handlers, state, configs, DB, etc.)
--- a/src/infrastructure/ollama.rs
+++ b/src/infrastructure/ollama.rs
@@ -0,0 +1,92 @@
 use dioxus::prelude::*;
 use serde::{Deserialize, Serialize};
 /// Status of a local Ollama instance, including connectivity and loaded models.
 ///
 /// # Fields
 ///
 /// * `online` - Whether the Ollama API responded successfully
 /// * `models` - List of model names currently available on the instance
 #[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
 pub struct OllamaStatus {
    pub online: bool,
    pub models: Vec<String>,
 }
 /// Response from Ollama's `GET /api/tags` endpoint.
 #[cfg(feature = "server")]
 #[derive(Deserialize)]
 struct OllamaTagsResponse {
    models: Vec<OllamaModel>,
 }
 /// A single model entry from Ollama's tags API.
 #[cfg(feature = "server")]
 #[derive(Deserialize)]
 struct OllamaModel {
    name: String,
 }
 /// Check the status of a local Ollama instance by querying its tags endpoint.
 ///
 /// Calls `GET <ollama_url>/api/tags` to list available models and determine
 /// whether the instance is reachable.
 ///
 /// # Arguments
 ///
 /// * `ollama_url` - Base URL of the Ollama instance (e.g. "http://localhost:11434")
 ///
 /// # Returns
 ///
 /// An `OllamaStatus` with `online: true` and model names if reachable,
 /// or `online: false` with an empty model list on failure
 ///
 /// # Errors
 ///
 /// Returns `ServerFnError` only on serialization issues; network failures
 /// are caught and returned as `online: false`
 #[post("/api/ollama-status")]
 pub async fn get_ollama_status(ollama_url: String) -> Result<OllamaStatus, ServerFnError> {
    let state: crate::infrastructure::ServerState =
        dioxus_fullstack::FullstackContext::extract().await?;
    let base_url = if ollama_url.is_empty() {
        state.services.ollama_url.clone()
    } else {
        ollama_url
    };
    let url = format!("{}/api/tags", base_url.trim_end_matches('/'));
    let client = reqwest::Client::builder()
        .timeout(std::time::Duration::from_secs(5))
        .build()
        .map_err(|e| ServerFnError::new(format!("HTTP client error: {e}")))?;
    let resp = match client.get(&url).send().await {
        Ok(r) if r.status().is_success() => r,
        _ => {
            return Ok(OllamaStatus {
                online: false,
                models: Vec::new(),
            });
        }
    };
    let body: OllamaTagsResponse = match resp.json().await {
        Ok(b) => b,
        Err(_) => {
            return Ok(OllamaStatus {
                online: true,
                models: Vec::new(),
            });
        }
    };
    let models = body.models.into_iter().map(|m| m.name).collect();
    Ok(OllamaStatus {
        online: true,
        models,
    })
 }
--- a/src/infrastructure/provider_client.rs
+++ b/src/infrastructure/provider_client.rs
@@ -1,6 +1,6 @@
 //! Unified LLM provider dispatch.
 //!
-//! Routes chat completion requests to LiteLLM, OpenAI, Anthropic, or
+//! Routes chat completion requests to Ollama, OpenAI, Anthropic, or
 //! HuggingFace based on the session's provider setting. All providers
 //! except Anthropic use the OpenAI-compatible chat completions format.
@@ -20,11 +20,11 @@ pub struct ProviderMessage {
 ///
 /// # Arguments
 ///
-/// * `state` - Server state (for default LiteLLM URL/model)
+/// * `state` - Server state (for default Ollama URL/model)
-/// * `provider` - Provider name (`"litellm"`, `"openai"`, `"anthropic"`, `"huggingface"`)
+/// * `provider` - Provider name (`"ollama"`, `"openai"`, `"anthropic"`, `"huggingface"`)
 /// * `model` - Model ID
 /// * `messages` - Conversation history
-/// * `api_key` - API key (required for non-LiteLLM providers; LiteLLM uses server config)
+/// * `api_key` - API key (required for non-Ollama providers)
 /// * `stream` - Whether to request streaming
 ///
 /// # Returns
@@ -123,11 +123,11 @@ pub async fn send_chat_request(
                .send()
                .await
        }
-        // Default: LiteLLM proxy (OpenAI-compatible endpoint)
+        // Default: Ollama (OpenAI-compatible endpoint)
        _ => {
-            let base_url = &state.services.litellm_url;
+            let base_url = &state.services.ollama_url;
            let resolved_model = if model.is_empty() {
-                &state.services.litellm_model
+                &state.services.ollama_model
            } else {
                model
            };
@@ -137,42 +137,12 @@ pub async fn send_chat_request(
                "messages": messages,
                "stream": stream,
            });
-            let litellm_key = &state.services.litellm_api_key;
+            client
            let mut request = client
                .post(&url)
                .header("content-type", "application/json")
-                .json(&body);
+                .json(&body)
-            if !litellm_key.is_empty() {
+                .send()
-                request = request.header("Authorization", format!("Bearer {litellm_key}"));
+                .await
            }
            request.send().await
        }
    }
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use pretty_assertions::assert_eq;
    #[test]
    fn provider_message_serde_round_trip() {
        let msg = ProviderMessage {
            role: "assistant".into(),
            content: "Hello, world!".into(),
        };
        let json = serde_json::to_string(&msg).expect("serialize ProviderMessage");
        let back: ProviderMessage =
            serde_json::from_str(&json).expect("deserialize ProviderMessage");
        assert_eq!(msg.role, back.role);
        assert_eq!(msg.content, back.content);
    }
    #[test]
    fn provider_message_deserialize_from_json_literal() {
        let json = r#"{"role":"user","content":"What is Rust?"}"#;
        let msg: ProviderMessage = serde_json::from_str(json).expect("deserialize from literal");
        assert_eq!(msg.role, "user");
        assert_eq!(msg.content, "What is Rust?");
    }
 }
--- a/src/infrastructure/searxng.rs
+++ b/src/infrastructure/searxng.rs
@@ -5,13 +5,13 @@ use dioxus::prelude::*;
 // The #[server] macro generates a client stub for the web build that
 // sends a network request instead of executing this function body.
 #[cfg(feature = "server")]
-pub(crate) mod inner {
+mod inner {
    use serde::Deserialize;
    use std::collections::HashSet;
    /// Individual result from the SearXNG search API.
    #[derive(Debug, Deserialize)]
-    pub(crate) struct SearxngResult {
+    pub(super) struct SearxngResult {
        pub title: String,
        pub url: String,
        pub content: Option<String>,
@@ -25,7 +25,7 @@ pub(crate) mod inner {
    /// Top-level response from the SearXNG search API.
    #[derive(Debug, Deserialize)]
-    pub(crate) struct SearxngResponse {
+    pub(super) struct SearxngResponse {
        pub results: Vec<SearxngResult>,
    }
@@ -40,7 +40,7 @@ pub(crate) mod inner {
    /// # Returns
    ///
    /// The domain host or a fallback "Web" string
-    pub(crate) fn extract_source(url_str: &str) -> String {
+    pub(super) fn extract_source(url_str: &str) -> String {
        url::Url::parse(url_str)
            .ok()
            .and_then(|u| u.host_str().map(String::from))
@@ -64,7 +64,7 @@ pub(crate) mod inner {
    /// # Returns
    ///
    /// Filtered, deduplicated, and ranked results
-    pub(crate) fn rank_and_deduplicate(
+    pub(super) fn rank_and_deduplicate(
        mut results: Vec<SearxngResult>,
        max_results: usize,
    ) -> Vec<SearxngResult> {
@@ -285,166 +285,3 @@ pub async fn get_trending_topics() -> Result<Vec<String>, ServerFnError> {
    Ok(topics)
 }
 #[cfg(all(test, feature = "server"))]
 mod tests {
    #![allow(clippy::unwrap_used, clippy::expect_used)]
    use super::inner::*;
    use pretty_assertions::assert_eq;
    // -----------------------------------------------------------------------
    // extract_source()
    // -----------------------------------------------------------------------
    #[test]
    fn extract_source_strips_www() {
        assert_eq!(
            extract_source("https://www.example.com/page"),
            "example.com"
        );
    }
    #[test]
    fn extract_source_returns_domain() {
        assert_eq!(
            extract_source("https://techcrunch.com/article"),
            "techcrunch.com"
        );
    }
    #[test]
    fn extract_source_invalid_url_returns_web() {
        assert_eq!(extract_source("not-a-url"), "Web");
    }
    #[test]
    fn extract_source_no_scheme_returns_web() {
        // url::Url::parse requires a scheme; bare domain fails
        assert_eq!(extract_source("example.com/path"), "Web");
    }
    // -----------------------------------------------------------------------
    // rank_and_deduplicate()
    // -----------------------------------------------------------------------
    fn make_result(url: &str, content: &str, score: f64) -> SearxngResult {
        SearxngResult {
            title: "Title".into(),
            url: url.into(),
            content: if content.is_empty() {
                None
            } else {
                Some(content.into())
            },
            published_date: None,
            thumbnail: None,
            score,
        }
    }
    #[test]
    fn rank_filters_empty_content() {
        let results = vec![
            make_result("https://a.com", "", 10.0),
            make_result(
                "https://b.com",
                "This is meaningful content that passes the length filter",
                5.0,
            ),
        ];
        let ranked = rank_and_deduplicate(results, 10);
        assert_eq!(ranked.len(), 1);
        assert_eq!(ranked[0].url, "https://b.com");
    }
    #[test]
    fn rank_filters_short_content() {
        let results = vec![
            make_result("https://a.com", "short", 10.0),
            make_result(
                "https://b.com",
                "This content is long enough to pass the 20-char filter threshold",
                5.0,
            ),
        ];
        let ranked = rank_and_deduplicate(results, 10);
        assert_eq!(ranked.len(), 1);
    }
    #[test]
    fn rank_deduplicates_by_domain_keeps_highest() {
        let results = vec![
            make_result(
                "https://example.com/page1",
                "First result with enough content here for the filter",
                3.0,
            ),
            make_result(
                "https://example.com/page2",
                "Second result with enough content here for the filter",
                8.0,
            ),
        ];
        let ranked = rank_and_deduplicate(results, 10);
        assert_eq!(ranked.len(), 1);
        // Should keep the highest-scored one (page2 with score 8.0)
        assert_eq!(ranked[0].url, "https://example.com/page2");
    }
    #[test]
    fn rank_sorts_by_score_descending() {
        let results = vec![
            make_result(
                "https://a.com/p",
                "Content A that is long enough to pass the filter check",
                1.0,
            ),
            make_result(
                "https://b.com/p",
                "Content B that is long enough to pass the filter check",
                5.0,
            ),
            make_result(
                "https://c.com/p",
                "Content C that is long enough to pass the filter check",
                3.0,
            ),
        ];
        let ranked = rank_and_deduplicate(results, 10);
        assert_eq!(ranked.len(), 3);
        assert!(ranked[0].score >= ranked[1].score);
        assert!(ranked[1].score >= ranked[2].score);
    }
    #[test]
    fn rank_truncates_to_max_results() {
        let results: Vec<_> = (0..20)
            .map(|i| {
                make_result(
                    &format!("https://site{i}.com/page"),
                    &format!("Content for site {i} that is long enough to pass the filter"),
                    i as f64,
                )
            })
            .collect();
        let ranked = rank_and_deduplicate(results, 5);
        assert_eq!(ranked.len(), 5);
    }
    #[test]
    fn rank_empty_input_returns_empty() {
        let ranked = rank_and_deduplicate(vec![], 10);
        assert!(ranked.is_empty());
    }
    #[test]
    fn rank_all_filtered_returns_empty() {
        let results = vec![
            make_result("https://a.com", "", 10.0),
            make_result("https://b.com", "too short", 5.0),
        ];
        let ranked = rank_and_deduplicate(results, 10);
        assert!(ranked.is_empty());
    }
 }
--- a/src/infrastructure/server_state.rs
+++ b/src/infrastructure/server_state.rs
@@ -45,7 +45,7 @@ pub struct ServerStateInner {
    pub keycloak: &'static KeycloakConfig,
    /// Outbound email settings.
    pub smtp: &'static SmtpConfig,
-    /// URLs for LiteLLM, SearXNG, LangChain, S3, etc.
+    /// URLs for Ollama, SearXNG, LangChain, S3, etc.
    pub services: &'static ServiceUrls,
    /// Stripe billing keys.
    pub stripe: &'static StripeConfig,
--- a/src/infrastructure/state.rs
+++ b/src/infrastructure/state.rs
@@ -44,91 +44,3 @@ pub struct User {
    /// Avatar / profile picture URL.
    pub avatar_url: String,
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use pretty_assertions::assert_eq;
    #[test]
    fn user_state_inner_default_has_empty_strings() {
        let inner = UserStateInner::default();
        assert_eq!(inner.sub, "");
        assert_eq!(inner.access_token, "");
        assert_eq!(inner.refresh_token, "");
        assert_eq!(inner.user.email, "");
        assert_eq!(inner.user.name, "");
        assert_eq!(inner.user.avatar_url, "");
    }
    #[test]
    fn user_default_has_empty_strings() {
        let user = User::default();
        assert_eq!(user.email, "");
        assert_eq!(user.name, "");
        assert_eq!(user.avatar_url, "");
    }
    #[test]
    fn user_state_inner_serde_round_trip() {
        let inner = UserStateInner {
            sub: "user-123".into(),
            access_token: "tok-abc".into(),
            refresh_token: "ref-xyz".into(),
            user: User {
                email: "a@b.com".into(),
                name: "Alice".into(),
                avatar_url: "https://img.example.com/a.png".into(),
            },
        };
        let json = serde_json::to_string(&inner).expect("serialize UserStateInner");
        let back: UserStateInner = serde_json::from_str(&json).expect("deserialize UserStateInner");
        assert_eq!(inner.sub, back.sub);
        assert_eq!(inner.access_token, back.access_token);
        assert_eq!(inner.refresh_token, back.refresh_token);
        assert_eq!(inner.user.email, back.user.email);
        assert_eq!(inner.user.name, back.user.name);
        assert_eq!(inner.user.avatar_url, back.user.avatar_url);
    }
    #[test]
    fn user_state_from_inner_and_deref() {
        let inner = UserStateInner {
            sub: "sub-1".into(),
            access_token: "at".into(),
            refresh_token: "rt".into(),
            user: User {
                email: "e@e.com".into(),
                name: "Eve".into(),
                avatar_url: "".into(),
            },
        };
        let state = UserState::from(inner);
        // Deref should give access to inner fields
        assert_eq!(state.sub, "sub-1");
        assert_eq!(state.user.name, "Eve");
    }
    #[test]
    fn user_serde_round_trip() {
        let user = User {
            email: "bob@test.com".into(),
            name: "Bob".into(),
            avatar_url: "https://avatars.io/bob".into(),
        };
        let json = serde_json::to_string(&user).expect("serialize User");
        let back: User = serde_json::from_str(&json).expect("deserialize User");
        assert_eq!(user.email, back.email);
        assert_eq!(user.name, back.name);
        assert_eq!(user.avatar_url, back.avatar_url);
    }
    #[test]
    fn user_state_clone_is_cheap() {
        let inner = UserStateInner::default();
        let state = UserState::from(inner);
        let cloned = state.clone();
        // Both point to the same Arc allocation
        assert_eq!(state.sub, cloned.sub);
    }
 }
--- a/src/models/chat.rs
+++ b/src/models/chat.rs
@@ -60,8 +60,8 @@ pub struct Attachment {
 /// * `user_sub` - Keycloak subject ID (session owner)
 /// * `title` - Display title (auto-generated or user-renamed)
 /// * `namespace` - Grouping for sidebar sections
-/// * `provider` - LLM provider used (e.g. "litellm", "openai")
+/// * `provider` - LLM provider used (e.g. "ollama", "openai")
-/// * `model` - Model ID used (e.g. "qwen3-32b")
+/// * `model` - Model ID used (e.g. "llama3.1:8b")
 /// * `created_at` - ISO 8601 creation timestamp
 /// * `updated_at` - ISO 8601 last-activity timestamp
 /// * `article_url` - Source article URL (for News namespace sessions)
@@ -105,163 +105,3 @@ pub struct ChatMessage {
    pub attachments: Vec<Attachment>,
    pub timestamp: String,
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use pretty_assertions::assert_eq;
    #[test]
    fn chat_namespace_default_is_general() {
        assert_eq!(ChatNamespace::default(), ChatNamespace::General);
    }
    #[test]
    fn chat_role_serde_round_trip() {
        for role in [ChatRole::User, ChatRole::Assistant, ChatRole::System] {
            let json =
                serde_json::to_string(&role).unwrap_or_else(|_| panic!("serialize {:?}", role));
            let back: ChatRole =
                serde_json::from_str(&json).unwrap_or_else(|_| panic!("deserialize {:?}", role));
            assert_eq!(role, back);
        }
    }
    #[test]
    fn chat_namespace_serde_round_trip() {
        for ns in [ChatNamespace::General, ChatNamespace::News] {
            let json = serde_json::to_string(&ns).unwrap_or_else(|_| panic!("serialize {:?}", ns));
            let back: ChatNamespace =
                serde_json::from_str(&json).unwrap_or_else(|_| panic!("deserialize {:?}", ns));
            assert_eq!(ns, back);
        }
    }
    #[test]
    fn attachment_kind_serde_round_trip() {
        for kind in [
            AttachmentKind::Image,
            AttachmentKind::Document,
            AttachmentKind::Code,
        ] {
            let json =
                serde_json::to_string(&kind).unwrap_or_else(|_| panic!("serialize {:?}", kind));
            let back: AttachmentKind =
                serde_json::from_str(&json).unwrap_or_else(|_| panic!("deserialize {:?}", kind));
            assert_eq!(kind, back);
        }
    }
    #[test]
    fn attachment_serde_round_trip() {
        let att = Attachment {
            name: "photo.png".into(),
            kind: AttachmentKind::Image,
            size_bytes: 2048,
        };
        let json = serde_json::to_string(&att).expect("serialize Attachment");
        let back: Attachment = serde_json::from_str(&json).expect("deserialize Attachment");
        assert_eq!(att, back);
    }
    #[test]
    fn chat_session_serde_round_trip() {
        let session = ChatSession {
            id: "abc123".into(),
            user_sub: "user-1".into(),
            title: "Test Chat".into(),
            namespace: ChatNamespace::General,
            provider: "litellm".into(),
            model: "qwen3-32b".into(),
            created_at: "2025-01-01T00:00:00Z".into(),
            updated_at: "2025-01-01T01:00:00Z".into(),
            article_url: None,
        };
        let json = serde_json::to_string(&session).expect("serialize ChatSession");
        let back: ChatSession = serde_json::from_str(&json).expect("deserialize ChatSession");
        assert_eq!(session, back);
    }
    #[test]
    fn chat_session_id_alias_deserialization() {
        // MongoDB returns `_id` instead of `id`
        let json = r#"{
            "_id": "mongo-id",
            "user_sub": "u1",
            "title": "t",
            "provider": "litellm",
            "model": "m",
            "created_at": "2025-01-01",
            "updated_at": "2025-01-01"
        }"#;
        let session: ChatSession = serde_json::from_str(json).expect("deserialize with _id");
        assert_eq!(session.id, "mongo-id");
    }
    #[test]
    fn chat_session_empty_id_skips_serialization() {
        let session = ChatSession {
            id: String::new(),
            user_sub: "u1".into(),
            title: "t".into(),
            namespace: ChatNamespace::default(),
            provider: "litellm".into(),
            model: "m".into(),
            created_at: "2025-01-01".into(),
            updated_at: "2025-01-01".into(),
            article_url: None,
        };
        let json = serde_json::to_string(&session).expect("serialize");
        // `id` field should be absent when empty due to skip_serializing_if
        assert!(!json.contains("\"id\""));
    }
    #[test]
    fn chat_session_none_article_url_skips_serialization() {
        let session = ChatSession {
            id: "s1".into(),
            user_sub: "u1".into(),
            title: "t".into(),
            namespace: ChatNamespace::default(),
            provider: "litellm".into(),
            model: "m".into(),
            created_at: "2025-01-01".into(),
            updated_at: "2025-01-01".into(),
            article_url: None,
        };
        let json = serde_json::to_string(&session).expect("serialize");
        assert!(!json.contains("article_url"));
    }
    #[test]
    fn chat_message_serde_round_trip() {
        let msg = ChatMessage {
            id: "msg-1".into(),
            session_id: "s1".into(),
            role: ChatRole::User,
            content: "Hello AI".into(),
            attachments: vec![Attachment {
                name: "doc.pdf".into(),
                kind: AttachmentKind::Document,
                size_bytes: 4096,
            }],
            timestamp: "2025-01-01T00:00:00Z".into(),
        };
        let json = serde_json::to_string(&msg).expect("serialize ChatMessage");
        let back: ChatMessage = serde_json::from_str(&json).expect("deserialize ChatMessage");
        assert_eq!(msg, back);
    }
    #[test]
    fn chat_message_id_alias_deserialization() {
        let json = r#"{
            "_id": "mongo-msg-id",
            "session_id": "s1",
            "role": "User",
            "content": "hi",
            "timestamp": "2025-01-01"
        }"#;
        let msg: ChatMessage = serde_json::from_str(json).expect("deserialize with _id");
        assert_eq!(msg.id, "mongo-msg-id");
    }
 }
--- a/src/models/developer.rs
+++ b/src/models/developer.rs
@@ -45,63 +45,3 @@ pub struct AnalyticsMetric {
    pub value: String,
    pub change_pct: f64,
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use pretty_assertions::assert_eq;
    #[test]
    fn agent_entry_serde_round_trip() {
        let agent = AgentEntry {
            id: "a1".into(),
            name: "RAG Agent".into(),
            description: "Retrieval-augmented generation".into(),
            status: "running".into(),
        };
        let json = serde_json::to_string(&agent).expect("serialize AgentEntry");
        let back: AgentEntry = serde_json::from_str(&json).expect("deserialize AgentEntry");
        assert_eq!(agent, back);
    }
    #[test]
    fn flow_entry_serde_round_trip() {
        let flow = FlowEntry {
            id: "f1".into(),
            name: "Data Pipeline".into(),
            node_count: 5,
            last_run: Some("2025-06-01T12:00:00Z".into()),
        };
        let json = serde_json::to_string(&flow).expect("serialize FlowEntry");
        let back: FlowEntry = serde_json::from_str(&json).expect("deserialize FlowEntry");
        assert_eq!(flow, back);
    }
    #[test]
    fn flow_entry_with_none_last_run() {
        let flow = FlowEntry {
            id: "f2".into(),
            name: "New Flow".into(),
            node_count: 0,
            last_run: None,
        };
        let json = serde_json::to_string(&flow).expect("serialize");
        let back: FlowEntry = serde_json::from_str(&json).expect("deserialize");
        assert_eq!(flow, back);
        assert_eq!(back.last_run, None);
    }
    #[test]
    fn analytics_metric_negative_change_pct() {
        let metric = AnalyticsMetric {
            label: "Latency".into(),
            value: "120ms".into(),
            change_pct: -15.5,
        };
        let json = serde_json::to_string(&metric).expect("serialize AnalyticsMetric");
        let back: AnalyticsMetric =
            serde_json::from_str(&json).expect("deserialize AnalyticsMetric");
        assert_eq!(metric, back);
        assert!(back.change_pct < 0.0);
    }
 }
--- a/src/models/mod.rs
+++ b/src/models/mod.rs
@@ -3,7 +3,6 @@ mod developer;
 mod news;
 mod organization;
 mod provider;
 mod services;
 mod user;
 pub use chat::*;
@@ -11,5 +10,4 @@ pub use developer::*;
 pub use news::*;
 pub use organization::*;
 pub use provider::*;
 pub use services::*;
 pub use user::*;
--- a/src/models/news.rs
+++ b/src/models/news.rs
@@ -23,61 +23,3 @@ pub struct NewsCard {
    pub thumbnail_url: Option<String>,
    pub published_at: String,
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use pretty_assertions::assert_eq;
    #[test]
    fn news_card_serde_round_trip() {
        let card = NewsCard {
            title: "AI Breakthrough".into(),
            source: "techcrunch.com".into(),
            summary: "New model released".into(),
            content: "Full article content here".into(),
            category: "AI".into(),
            url: "https://example.com/article".into(),
            thumbnail_url: Some("https://example.com/thumb.jpg".into()),
            published_at: "2025-06-01".into(),
        };
        let json = serde_json::to_string(&card).expect("serialize NewsCard");
        let back: NewsCard = serde_json::from_str(&json).expect("deserialize NewsCard");
        assert_eq!(card, back);
    }
    #[test]
    fn news_card_thumbnail_none() {
        let card = NewsCard {
            title: "No Thumb".into(),
            source: "bbc.com".into(),
            summary: "Summary".into(),
            content: "Content".into(),
            category: "Tech".into(),
            url: "https://bbc.com/article".into(),
            thumbnail_url: None,
            published_at: "2025-06-01".into(),
        };
        let json = serde_json::to_string(&card).expect("serialize");
        let back: NewsCard = serde_json::from_str(&json).expect("deserialize");
        assert_eq!(card, back);
    }
    #[test]
    fn news_card_thumbnail_some() {
        let card = NewsCard {
            title: "With Thumb".into(),
            source: "cnn.com".into(),
            summary: "Summary".into(),
            content: "Content".into(),
            category: "News".into(),
            url: "https://cnn.com/article".into(),
            thumbnail_url: Some("https://cnn.com/img.jpg".into()),
            published_at: "2025-06-01".into(),
        };
        let json = serde_json::to_string(&card).expect("serialize");
        assert!(json.contains("img.jpg"));
        let back: NewsCard = serde_json::from_str(&json).expect("deserialize");
        assert_eq!(card.thumbnail_url, back.thumbnail_url);
    }
 }
--- a/src/models/organization.rs
+++ b/src/models/organization.rs
@@ -83,42 +83,6 @@ pub struct BillingUsage {
    pub billing_cycle_end: String,
 }
 /// Aggregated token usage statistics from LiteLLM's spend tracking API.
 ///
 /// # Fields
 ///
 /// * `total_spend` - Total cost in USD across all models
 /// * `total_prompt_tokens` - Sum of prompt (input) tokens
 /// * `total_completion_tokens` - Sum of completion (output) tokens
 /// * `total_tokens` - Sum of all tokens (prompt + completion)
 /// * `model_breakdown` - Per-model usage breakdown
 #[derive(Debug, Clone, Default, PartialEq, Serialize, Deserialize)]
 pub struct LitellmUsageStats {
    pub total_spend: f64,
    pub total_prompt_tokens: u64,
    pub total_completion_tokens: u64,
    pub total_tokens: u64,
    pub model_breakdown: Vec<ModelUsage>,
 }
 /// Token and spend usage for a single LLM model.
 ///
 /// # Fields
 ///
 /// * `model` - Model identifier (e.g. "gpt-4", "claude-3-opus")
 /// * `spend` - Cost in USD for this model
 /// * `prompt_tokens` - Prompt (input) tokens consumed
 /// * `completion_tokens` - Completion (output) tokens generated
 /// * `total_tokens` - Total tokens (prompt + completion)
 #[derive(Debug, Clone, Default, PartialEq, Serialize, Deserialize)]
 pub struct ModelUsage {
    pub model: String,
    pub spend: f64,
    pub prompt_tokens: u64,
    pub completion_tokens: u64,
    pub total_tokens: u64,
 }
 /// Organisation-level settings stored in MongoDB.
 ///
 /// These complement Keycloak's Organizations feature with
@@ -152,200 +116,3 @@ pub struct OrgBillingRecord {
    /// Number of tokens consumed during this cycle.
    pub tokens_used: u64,
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use pretty_assertions::assert_eq;
    #[test]
    fn member_role_label_admin() {
        assert_eq!(MemberRole::Admin.label(), "Admin");
    }
    #[test]
    fn member_role_label_member() {
        assert_eq!(MemberRole::Member.label(), "Member");
    }
    #[test]
    fn member_role_label_viewer() {
        assert_eq!(MemberRole::Viewer.label(), "Viewer");
    }
    #[test]
    fn member_role_all_returns_three_in_order() {
        let all = MemberRole::all();
        assert_eq!(all.len(), 3);
        assert_eq!(all[0], MemberRole::Admin);
        assert_eq!(all[1], MemberRole::Member);
        assert_eq!(all[2], MemberRole::Viewer);
    }
    #[test]
    fn member_role_serde_round_trip() {
        for role in MemberRole::all() {
            let json =
                serde_json::to_string(role).unwrap_or_else(|_| panic!("serialize {:?}", role));
            let back: MemberRole =
                serde_json::from_str(&json).unwrap_or_else(|_| panic!("deserialize {:?}", role));
            assert_eq!(*role, back);
        }
    }
    #[test]
    fn org_member_serde_round_trip() {
        let member = OrgMember {
            id: "m1".into(),
            name: "Alice".into(),
            email: "alice@example.com".into(),
            role: MemberRole::Admin,
            joined_at: "2025-01-01T00:00:00Z".into(),
        };
        let json = serde_json::to_string(&member).expect("serialize OrgMember");
        let back: OrgMember = serde_json::from_str(&json).expect("deserialize OrgMember");
        assert_eq!(member, back);
    }
    #[test]
    fn pricing_plan_with_max_seats() {
        let plan = PricingPlan {
            id: "team".into(),
            name: "Team".into(),
            price_eur: 49,
            features: vec!["SSO".into(), "Priority".into()],
            highlighted: true,
            max_seats: Some(25),
        };
        let json = serde_json::to_string(&plan).expect("serialize PricingPlan");
        let back: PricingPlan = serde_json::from_str(&json).expect("deserialize PricingPlan");
        assert_eq!(plan, back);
    }
    #[test]
    fn pricing_plan_without_max_seats() {
        let plan = PricingPlan {
            id: "enterprise".into(),
            name: "Enterprise".into(),
            price_eur: 199,
            features: vec!["Unlimited".into()],
            highlighted: false,
            max_seats: None,
        };
        let json = serde_json::to_string(&plan).expect("serialize PricingPlan");
        let back: PricingPlan = serde_json::from_str(&json).expect("deserialize PricingPlan");
        assert_eq!(plan, back);
        assert!(json.contains("null") || !json.contains("max_seats"));
    }
    #[test]
    fn billing_usage_serde_round_trip() {
        let usage = BillingUsage {
            seats_used: 5,
            seats_total: 10,
            tokens_used: 1_000_000,
            tokens_limit: 5_000_000,
            billing_cycle_end: "2025-12-31".into(),
        };
        let json = serde_json::to_string(&usage).expect("serialize BillingUsage");
        let back: BillingUsage = serde_json::from_str(&json).expect("deserialize BillingUsage");
        assert_eq!(usage, back);
    }
    #[test]
    fn org_settings_default() {
        let settings = OrgSettings::default();
        assert_eq!(settings.org_id, "");
        assert_eq!(settings.plan_id, "");
        assert!(settings.enabled_features.is_empty());
        assert_eq!(settings.stripe_customer_id, "");
    }
    #[test]
    fn org_billing_record_default() {
        let record = OrgBillingRecord::default();
        assert_eq!(record.org_id, "");
        assert_eq!(record.cycle_start, "");
        assert_eq!(record.cycle_end, "");
        assert_eq!(record.seats_used, 0);
        assert_eq!(record.tokens_used, 0);
    }
    #[test]
    fn litellm_usage_stats_default() {
        let stats = LitellmUsageStats::default();
        assert_eq!(stats.total_spend, 0.0);
        assert_eq!(stats.total_prompt_tokens, 0);
        assert_eq!(stats.total_completion_tokens, 0);
        assert_eq!(stats.total_tokens, 0);
        assert!(stats.model_breakdown.is_empty());
    }
    #[test]
    fn litellm_usage_stats_serde_round_trip() {
        let stats = LitellmUsageStats {
            total_spend: 12.34,
            total_prompt_tokens: 50_000,
            total_completion_tokens: 25_000,
            total_tokens: 75_000,
            model_breakdown: vec![
                ModelUsage {
                    model: "gpt-4".into(),
                    spend: 10.0,
                    prompt_tokens: 40_000,
                    completion_tokens: 20_000,
                    total_tokens: 60_000,
                },
                ModelUsage {
                    model: "claude-3-opus".into(),
                    spend: 2.34,
                    prompt_tokens: 10_000,
                    completion_tokens: 5_000,
                    total_tokens: 15_000,
                },
            ],
        };
        let json = serde_json::to_string(&stats).expect("serialize LitellmUsageStats");
        let back: LitellmUsageStats =
            serde_json::from_str(&json).expect("deserialize LitellmUsageStats");
        assert_eq!(stats, back);
    }
    #[test]
    fn model_usage_default() {
        let usage = ModelUsage::default();
        assert_eq!(usage.model, "");
        assert_eq!(usage.spend, 0.0);
        assert_eq!(usage.prompt_tokens, 0);
        assert_eq!(usage.completion_tokens, 0);
        assert_eq!(usage.total_tokens, 0);
    }
    #[test]
    fn model_usage_serde_round_trip() {
        let usage = ModelUsage {
            model: "gpt-4-turbo".into(),
            spend: 5.67,
            prompt_tokens: 30_000,
            completion_tokens: 15_000,
            total_tokens: 45_000,
        };
        let json = serde_json::to_string(&usage).expect("serialize ModelUsage");
        let back: ModelUsage = serde_json::from_str(&json).expect("deserialize ModelUsage");
        assert_eq!(usage, back);
    }
    #[test]
    fn litellm_usage_stats_empty_breakdown_round_trip() {
        let stats = LitellmUsageStats {
            total_spend: 0.0,
            total_prompt_tokens: 0,
            total_completion_tokens: 0,
            total_tokens: 0,
            model_breakdown: Vec::new(),
        };
        let json = serde_json::to_string(&stats).expect("serialize empty stats");
        let back: LitellmUsageStats = serde_json::from_str(&json).expect("deserialize empty stats");
        assert_eq!(stats, back);
    }
 }
--- a/src/models/provider.rs
+++ b/src/models/provider.rs
@@ -3,8 +3,8 @@ use serde::{Deserialize, Serialize};
 /// Supported LLM provider backends.
 #[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
 pub enum LlmProvider {
-    /// LiteLLM proxy for unified model access
+    /// Self-hosted models via Ollama
-    LiteLlm,
+    Ollama,
    /// Hugging Face Inference API
    HuggingFace,
    /// OpenAI-compatible endpoints
@@ -17,7 +17,7 @@ impl LlmProvider {
    /// Returns the display name for a provider.
    pub fn label(&self) -> &'static str {
        match self {
-            Self::LiteLlm => "LiteLLM",
+            Self::Ollama => "Ollama",
            Self::HuggingFace => "Hugging Face",
            Self::OpenAi => "OpenAI",
            Self::Anthropic => "Anthropic",
@@ -29,7 +29,7 @@ impl LlmProvider {
 ///
 /// # Fields
 ///
-/// * `id` - Unique model identifier (e.g. "qwen3-32b")
+/// * `id` - Unique model identifier (e.g. "llama3.1:8b")
 /// * `name` - Human-readable display name
 /// * `provider` - Which provider hosts this model
 /// * `context_window` - Maximum context length in tokens
@@ -72,84 +72,3 @@ pub struct ProviderConfig {
    pub selected_embedding: String,
    pub api_key_set: bool,
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use pretty_assertions::assert_eq;
    #[test]
    fn llm_provider_label_litellm() {
        assert_eq!(LlmProvider::LiteLlm.label(), "LiteLLM");
    }
    #[test]
    fn llm_provider_label_hugging_face() {
        assert_eq!(LlmProvider::HuggingFace.label(), "Hugging Face");
    }
    #[test]
    fn llm_provider_label_openai() {
        assert_eq!(LlmProvider::OpenAi.label(), "OpenAI");
    }
    #[test]
    fn llm_provider_label_anthropic() {
        assert_eq!(LlmProvider::Anthropic.label(), "Anthropic");
    }
    #[test]
    fn llm_provider_serde_round_trip() {
        for variant in [
            LlmProvider::LiteLlm,
            LlmProvider::HuggingFace,
            LlmProvider::OpenAi,
            LlmProvider::Anthropic,
        ] {
            let json = serde_json::to_string(&variant)
                .unwrap_or_else(|_| panic!("serialize {:?}", variant));
            let back: LlmProvider =
                serde_json::from_str(&json).unwrap_or_else(|_| panic!("deserialize {:?}", variant));
            assert_eq!(variant, back);
        }
    }
    #[test]
    fn model_entry_serde_round_trip() {
        let entry = ModelEntry {
            id: "qwen3-32b".into(),
            name: "Qwen3 32B".into(),
            provider: LlmProvider::LiteLlm,
            context_window: 32,
        };
        let json = serde_json::to_string(&entry).expect("serialize ModelEntry");
        let back: ModelEntry = serde_json::from_str(&json).expect("deserialize ModelEntry");
        assert_eq!(entry, back);
    }
    #[test]
    fn embedding_entry_serde_round_trip() {
        let entry = EmbeddingEntry {
            id: "nomic-embed".into(),
            name: "Nomic Embed".into(),
            provider: LlmProvider::HuggingFace,
            dimensions: 768,
        };
        let json = serde_json::to_string(&entry).expect("serialize EmbeddingEntry");
        let back: EmbeddingEntry = serde_json::from_str(&json).expect("deserialize EmbeddingEntry");
        assert_eq!(entry, back);
    }
    #[test]
    fn provider_config_serde_round_trip() {
        let cfg = ProviderConfig {
            provider: LlmProvider::Anthropic,
            selected_model: "claude-3".into(),
            selected_embedding: "embed-v1".into(),
            api_key_set: true,
        };
        let json = serde_json::to_string(&cfg).expect("serialize ProviderConfig");
        let back: ProviderConfig = serde_json::from_str(&json).expect("deserialize ProviderConfig");
        assert_eq!(cfg, back);
    }
 }
--- a/src/models/services.rs
+++ b/src/models/services.rs
@@ -1,43 +0,0 @@
 use serde::{Deserialize, Serialize};
 /// Frontend-facing URLs for developer tool services.
 ///
 /// Provided as a context signal in `AppShell` so that developer pages
 /// can read the configured URLs without threading props through layouts.
 /// An empty string indicates the service is not configured.
 #[derive(Debug, Clone, Default, PartialEq, Serialize, Deserialize)]
 pub struct ServiceUrlsContext {
    /// LangGraph agent builder URL (empty if not configured)
    pub langgraph_url: String,
    /// LangFlow visual workflow builder URL (empty if not configured)
    pub langflow_url: String,
    /// Langfuse observability URL (empty if not configured)
    pub langfuse_url: String,
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use pretty_assertions::assert_eq;
    #[test]
    fn default_urls_are_empty() {
        let ctx = ServiceUrlsContext::default();
        assert_eq!(ctx.langgraph_url, "");
        assert_eq!(ctx.langflow_url, "");
        assert_eq!(ctx.langfuse_url, "");
    }
    #[test]
    fn serde_round_trip() {
        let ctx = ServiceUrlsContext {
            langgraph_url: "http://localhost:8123".into(),
            langflow_url: "http://localhost:7860".into(),
            langfuse_url: "http://localhost:3000".into(),
        };
        let json = serde_json::to_string(&ctx).expect("serialize ServiceUrlsContext");
        let back: ServiceUrlsContext =
            serde_json::from_str(&json).expect("deserialize ServiceUrlsContext");
        assert_eq!(ctx, back);
    }
 }
--- a/src/models/user.rs
+++ b/src/models/user.rs
@@ -24,23 +24,17 @@ pub struct AuthInfo {
    pub avatar_url: String,
    /// LibreChat instance URL for the sidebar chat link
    pub librechat_url: String,
    /// LangGraph agent builder URL (empty if not configured)
    pub langgraph_url: String,
    /// LangFlow visual workflow builder URL (empty if not configured)
    pub langflow_url: String,
    /// Langfuse observability URL (empty if not configured)
    pub langfuse_url: String,
 }
 /// Per-user LLM provider configuration stored in MongoDB.
 ///
 /// Controls which provider and model the user's chat sessions default
-/// to, and stores API keys for non-LiteLLM providers.
+/// to, and stores API keys for non-Ollama providers.
 #[derive(Debug, Clone, Default, PartialEq, Serialize, Deserialize)]
 pub struct UserProviderConfig {
-    /// Default provider name (e.g. "litellm", "openai")
+    /// Default provider name (e.g. "ollama", "openai")
    pub default_provider: String,
-    /// Default model ID (e.g. "qwen3-32b", "gpt-4o")
+    /// Default model ID (e.g. "llama3.1:8b", "gpt-4o")
    pub default_model: String,
    /// OpenAI API key (empty if not configured)
    #[serde(default, skip_serializing_if = "Option::is_none")]
@@ -51,8 +45,8 @@ pub struct UserProviderConfig {
    /// HuggingFace API key
    #[serde(default, skip_serializing_if = "Option::is_none")]
    pub huggingface_api_key: Option<String>,
-    /// Custom LiteLLM URL override (empty = use server default)
+    /// Custom Ollama URL override (empty = use server default)
-    pub litellm_url_override: String,
+    pub ollama_url_override: String,
 }
 /// Per-user preferences stored in MongoDB.
@@ -66,97 +60,13 @@ pub struct UserPreferences {
    pub org_id: String,
    /// User-selected news/search topics
    pub custom_topics: Vec<String>,
-    /// Per-user LiteLLM URL override (empty = use server default)
+    /// Per-user Ollama URL override (empty = use server default)
-    pub litellm_url_override: String,
+    pub ollama_url_override: String,
-    /// Per-user LiteLLM model override (empty = use server default)
+    /// Per-user Ollama model override (empty = use server default)
-    pub litellm_model_override: String,
+    pub ollama_model_override: String,
    /// Recently searched queries for quick access
    pub recent_searches: Vec<String>,
    /// LLM provider configuration
    #[serde(default)]
    pub provider_config: UserProviderConfig,
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use pretty_assertions::assert_eq;
    #[test]
    fn user_data_default() {
        let ud = UserData::default();
        assert_eq!(ud.name, "");
    }
    #[test]
    fn auth_info_default_not_authenticated() {
        let info = AuthInfo::default();
        assert!(!info.authenticated);
        assert_eq!(info.sub, "");
        assert_eq!(info.email, "");
        assert_eq!(info.name, "");
        assert_eq!(info.avatar_url, "");
        assert_eq!(info.librechat_url, "");
        assert_eq!(info.langgraph_url, "");
        assert_eq!(info.langflow_url, "");
        assert_eq!(info.langfuse_url, "");
    }
    #[test]
    fn auth_info_serde_round_trip() {
        let info = AuthInfo {
            authenticated: true,
            sub: "sub-123".into(),
            email: "test@example.com".into(),
            name: "Test User".into(),
            avatar_url: "https://example.com/avatar.png".into(),
            librechat_url: "https://chat.example.com".into(),
            langgraph_url: "http://localhost:8123".into(),
            langflow_url: "http://localhost:7860".into(),
            langfuse_url: "http://localhost:3000".into(),
        };
        let json = serde_json::to_string(&info).expect("serialize AuthInfo");
        let back: AuthInfo = serde_json::from_str(&json).expect("deserialize AuthInfo");
        assert_eq!(info, back);
    }
    #[test]
    fn user_preferences_default() {
        let prefs = UserPreferences::default();
        assert_eq!(prefs.sub, "");
        assert_eq!(prefs.org_id, "");
        assert!(prefs.custom_topics.is_empty());
        assert!(prefs.recent_searches.is_empty());
    }
    #[test]
    fn user_provider_config_optional_keys_skip_none() {
        let cfg = UserProviderConfig {
            default_provider: "litellm".into(),
            default_model: "qwen3-32b".into(),
            openai_api_key: None,
            anthropic_api_key: None,
            huggingface_api_key: None,
            litellm_url_override: String::new(),
        };
        let json = serde_json::to_string(&cfg).expect("serialize UserProviderConfig");
        assert!(!json.contains("openai_api_key"));
        assert!(!json.contains("anthropic_api_key"));
        assert!(!json.contains("huggingface_api_key"));
    }
    #[test]
    fn user_provider_config_serde_round_trip_with_keys() {
        let cfg = UserProviderConfig {
            default_provider: "openai".into(),
            default_model: "gpt-4o".into(),
            openai_api_key: Some("sk-test".into()),
            anthropic_api_key: Some("ak-test".into()),
            huggingface_api_key: None,
            litellm_url_override: "http://custom:4000".into(),
        };
        let json = serde_json::to_string(&cfg).expect("serialize");
        let back: UserProviderConfig = serde_json::from_str(&json).expect("deserialize");
        assert_eq!(cfg, back);
    }
 }
--- a/src/pages/dashboard.rs
+++ b/src/pages/dashboard.rs
@@ -25,8 +25,8 @@ const DEFAULT_TOPICS: &[&str] = &[
 ///
 /// State is persisted across sessions using localStorage:
 /// - `certifai_topics`: custom user-defined search topics
-/// - `certifai_litellm_url`: LiteLLM proxy URL for summarization
+/// - `certifai_ollama_url`: Ollama instance URL for summarization
-/// - `certifai_litellm_model`: LiteLLM model ID for summarization
+/// - `certifai_ollama_model`: Ollama model ID for summarization
 #[component]
 pub fn DashboardPage() -> Element {
    let locale = use_context::<Signal<Locale>>();
@@ -34,11 +34,11 @@ pub fn DashboardPage() -> Element {
    // Persistent state stored in localStorage
    let mut custom_topics = use_persistent("certifai_topics".to_string(), Vec::<String>::new);
-    // Default to empty so the server functions use LITELLM_URL / LITELLM_MODEL
+    // Default to empty so the server functions use OLLAMA_URL / OLLAMA_MODEL
    // from .env. Only stores a non-empty value when the user explicitly saves
    // an override via the Settings panel.
-    let mut litellm_url = use_persistent("certifai_litellm_url".to_string(), String::new);
+    let mut ollama_url = use_persistent("certifai_ollama_url".to_string(), String::new);
-    let mut litellm_model = use_persistent("certifai_litellm_model".to_string(), String::new);
+    let mut ollama_model = use_persistent("certifai_ollama_model".to_string(), String::new);
    // Reactive signals for UI state
    let mut active_topic = use_signal(|| "AI".to_string());
@@ -235,8 +235,8 @@ pub fn DashboardPage() -> Element {
                    onclick: move |_| {
                        let currently_shown = *show_settings.read();
                        if !currently_shown {
-                            settings_url.set(litellm_url.read().clone());
+                            settings_url.set(ollama_url.read().clone());
-                            settings_model.set(litellm_model.read().clone());
+                            settings_model.set(ollama_model.read().clone());
                        }
                        show_settings.set(!currently_shown);
                    },
@@ -247,16 +247,16 @@ pub fn DashboardPage() -> Element {
            // Settings panel (collapsible)
            if *show_settings.read() {
                div { class: "settings-panel",
-                    h4 { class: "settings-panel-title", "{t(l, \"dashboard.litellm_settings\")}" }
+                    h4 { class: "settings-panel-title", "{t(l, \"dashboard.ollama_settings\")}" }
                    p { class: "settings-hint",
                        "{t(l, \"dashboard.settings_hint\")}"
                    }
                    div { class: "settings-field",
-                        label { "{t(l, \"dashboard.litellm_url\")}" }
+                        label { "{t(l, \"dashboard.ollama_url\")}" }
                        input {
                            class: "settings-input",
                            r#type: "text",
-                            placeholder: "{t(l, \"dashboard.litellm_url_placeholder\")}",
+                            placeholder: "{t(l, \"dashboard.ollama_url_placeholder\")}",
                            value: "{settings_url}",
                            oninput: move |e| settings_url.set(e.value()),
                        }
@@ -274,8 +274,8 @@ pub fn DashboardPage() -> Element {
                    button {
                        class: "btn btn-primary",
                        onclick: move |_| {
-                            *litellm_url.write() = settings_url.read().trim().to_string();
+                            *ollama_url.write() = settings_url.read().trim().to_string();
-                            *litellm_model.write() = settings_model.read().trim().to_string();
+                            *ollama_model.write() = settings_model.read().trim().to_string();
                            show_settings.set(false);
                        },
                        "{t(l, \"common.save\")}"
@@ -320,14 +320,14 @@ pub fn DashboardPage() -> Element {
                                            news_session_id.set(None);
-                                            let ll_url = litellm_url.read().clone();
+                                            let oll_url = ollama_url.read().clone();
-                                            let mdl = litellm_model.read().clone();
+                                            let mdl = ollama_model.read().clone();
                                            spawn(async move {
                                                is_summarizing.set(true);
                                                match crate::infrastructure::llm::summarize_article(
                                                        snippet.clone(),
                                                        article_url,
-                                                        ll_url,
+                                                        oll_url,
                                                        mdl,
                                                    )
                                                    .await
@@ -373,8 +373,8 @@ pub fn DashboardPage() -> Element {
                            chat_messages: chat_messages.read().clone(),
                            is_chatting: *is_chatting.read(),
                            on_chat_send: move |question: String| {
-                                let ll_url = litellm_url.read().clone();
+                                let oll_url = ollama_url.read().clone();
-                                let mdl = litellm_model.read().clone();
+                                let mdl = ollama_model.read().clone();
                                let ctx = article_context.read().clone();
                                // Capture article info for News session creation
                                let card_title = selected_card
@@ -394,7 +394,7 @@ pub fn DashboardPage() -> Element {
                                    content: question.clone(),
                                });
-                                // Build full message history for LiteLLM
+                                // Build full message history for Ollama
                                let system_msg = format!(
                                    "You are a helpful assistant. The user is reading \
                                     a news article. Use the following context to answer \
@@ -422,7 +422,7 @@ pub fn DashboardPage() -> Element {
                                        match create_chat_session(
                                            card_title,
                                            "News".to_string(),
-                                            "litellm".to_string(),
+                                            "ollama".to_string(),
                                            mdl.clone(),
                                            card_url,
                                        )
@@ -458,7 +458,7 @@ pub fn DashboardPage() -> Element {
                                    }
                                    match crate::infrastructure::llm::chat_followup(
-                                        msgs, ll_url, mdl,
+                                        msgs, oll_url, mdl,
                                    )
                                    .await
                                    {
@@ -495,7 +495,7 @@ pub fn DashboardPage() -> Element {
                // Right: sidebar (when no card selected)
                if !has_selection {
                    DashboardSidebar {
-                        litellm_url: litellm_url.read().clone(),
+                        ollama_url: ollama_url.read().clone(),
                        trending: trending_topics.clone(),
                        recent_searches: recent_searches.read().clone(),
                        on_topic_click: move |topic: String| {
--- a/src/pages/developer/agents.rs
+++ b/src/pages/developer/agents.rs
@@ -1,239 +1,26 @@
 use dioxus::prelude::*;
 use dioxus_free_icons::icons::bs_icons::{
    BsBook, BsBoxArrowUpRight, BsCodeSquare, BsCpu, BsGithub, BsLightningCharge,
 };
 use dioxus_free_icons::Icon;
 use crate::i18n::{t, Locale};
 use crate::models::ServiceUrlsContext;
-/// Agents informational landing page for LangGraph.
+/// Agents page placeholder for the LangGraph agent builder.
 ///
-/// Since LangGraph is API-only (no web UI), this page displays a hero section
+/// Shows a "Coming Soon" card with a disabled launch button.
-/// explaining its role, a connection status indicator, a card grid linking
+/// Will eventually integrate with the LangGraph framework.
 /// to documentation, and a live table of registered agents fetched from the
 /// LangGraph assistants API.
 #[component]
 pub fn AgentsPage() -> Element {
    let locale = use_context::<Signal<Locale>>();
    let svc = use_context::<Signal<ServiceUrlsContext>>();
    let l = *locale.read();
    let url = svc.read().langgraph_url.clone();
    // Derive whether a LangGraph URL is configured
    let connected = !url.is_empty();
    // Build the API reference URL from the configured base, falling back to "#"
    let api_ref_href = if connected {
        format!("{}/docs", url)
    } else {
        "#".to_string()
    };
    // Fetch agents from LangGraph when connected
    let agents_resource = use_resource(move || async move {
        match crate::infrastructure::langgraph::list_langgraph_agents().await {
            Ok(agents) => agents,
            Err(e) => {
                tracing::error!("Failed to fetch agents: {e}");
                Vec::new()
            }
        }
    });
    rsx! {
-        div { class: "agents-page",
+        section { class: "placeholder-page",
-            // -- Hero section --
+            div { class: "placeholder-card",
-            div { class: "agents-hero",
+                div { class: "placeholder-icon", "A" }
-                div { class: "agents-hero-row",
+                h2 { "{t(l, \"developer.agents_title\")}" }
-                    div { class: "agents-hero-icon",
+                p { class: "placeholder-desc",
-                        Icon { icon: BsCpu, width: 24, height: 24 }
+                    "{t(l, \"developer.agents_desc\")}"
                    }
                    h2 { class: "agents-hero-title",
                        {t(l, "developer.agents_title")}
                    }
                }
                p { class: "agents-hero-desc",
                    {t(l, "developer.agents_desc")}
                }
                // -- Connection status --
                if connected {
                    div { class: "agents-status",
                        span {
                            class: "agents-status-dot agents-status-dot--on",
                        }
                        span { {t(l, "developer.agents_status_connected")} }
                        code { class: "agents-status-url", {url.clone()} }
                    }
                } else {
                    div { class: "agents-status",
                        span {
                            class: "agents-status-dot agents-status-dot--off",
                        }
                        span { {t(l, "developer.agents_status_not_connected")} }
                        span { class: "agents-status-hint",
                            {t(l, "developer.agents_config_hint")}
                        }
                    }
                }
            }
            // -- Running Agents table --
            div { class: "agents-table-section",
                h3 { class: "agents-section-title",
                    {t(l, "developer.agents_running_title")}
                }
                match agents_resource.read().as_ref() {
                    None => {
                        rsx! {
                            p { class: "agents-table-loading",
                                {t(l, "common.loading")}
                            }
                        }
                    }
                    Some(agents) if agents.is_empty() => {
                        rsx! {
                            p { class: "agents-table-empty",
                                {t(l, "developer.agents_none")}
                            }
                        }
                    }
                    Some(agents) => {
                        rsx! {
                            div { class: "agents-table-wrap",
                                table { class: "agents-table",
                                    thead {
                                        tr {
                                            th { {t(l, "developer.agents_col_name")} }
                                            th { {t(l, "developer.agents_col_id")} }
                                            th { {t(l, "developer.agents_col_description")} }
                                            th { {t(l, "developer.agents_col_status")} }
                                        }
                                    }
                                    tbody {
                                        for agent in agents.iter() {
                                            tr { key: "{agent.id}",
                                                td { class: "agents-cell-name",
                                                    {agent.name.clone()}
                                                }
                                                td {
                                                    code { class: "agents-cell-id",
                                                        {agent.id.clone()}
                                                    }
                                                }
                                                td { class: "agents-cell-desc",
                                                    if agent.description.is_empty() {
                                                        span { class: "agents-cell-none", "--" }
                                                    } else {
                                                        {agent.description.clone()}
                                                    }
                                                }
                                                td {
                                                    span { class: "agents-badge agents-badge--active",
                                                        {agent.status.clone()}
                                                    }
                                                }
                                            }
                                        }
                                    }
                                }
                            }
                        }
                    }
                }
            }
            // -- Quick Start card grid --
            h3 { class: "agents-section-title",
                {t(l, "developer.agents_quick_start")}
            }
            div { class: "agents-grid",
                // Documentation
                a {
                    class: "agents-card",
                    href: "https://langchain-ai.github.io/langgraph/",
                    target: "_blank",
                    rel: "noopener noreferrer",
                    div { class: "agents-card-icon",
                        Icon { icon: BsBook, width: 18, height: 18 }
                    }
                    div { class: "agents-card-title",
                        {t(l, "developer.agents_docs")}
                    }
                    div { class: "agents-card-desc",
                        {t(l, "developer.agents_docs_desc")}
                    }
                }
                // Getting Started
                a {
                    class: "agents-card",
                    href: "https://langchain-ai.github.io/langgraph/tutorials/introduction/",
                    target: "_blank",
                    rel: "noopener noreferrer",
                    div { class: "agents-card-icon",
                        Icon { icon: BsLightningCharge, width: 18, height: 18 }
                    }
                    div { class: "agents-card-title",
                        {t(l, "developer.agents_getting_started")}
                    }
                    div { class: "agents-card-desc",
                        {t(l, "developer.agents_getting_started_desc")}
                    }
                }
                // GitHub
                a {
                    class: "agents-card",
                    href: "https://github.com/langchain-ai/langgraph",
                    target: "_blank",
                    rel: "noopener noreferrer",
                    div { class: "agents-card-icon",
                        Icon { icon: BsGithub, width: 18, height: 18 }
                    }
                    div { class: "agents-card-title",
                        {t(l, "developer.agents_github")}
                    }
                    div { class: "agents-card-desc",
                        {t(l, "developer.agents_github_desc")}
                    }
                }
                // Examples
                a {
                    class: "agents-card",
                    href: "https://github.com/langchain-ai/langgraph/tree/main/examples",
                    target: "_blank",
                    rel: "noopener noreferrer",
                    div { class: "agents-card-icon",
                        Icon { icon: BsCodeSquare, width: 18, height: 18 }
                    }
                    div { class: "agents-card-title",
                        {t(l, "developer.agents_examples")}
                    }
                    div { class: "agents-card-desc",
                        {t(l, "developer.agents_examples_desc")}
                    }
                }
                // API Reference (disabled when URL is empty)
                a {
                    class: if connected { "agents-card" } else { "agents-card agents-card--disabled" },
                    href: "{api_ref_href}",
                    target: "_blank",
                    rel: "noopener noreferrer",
                    div { class: "agents-card-icon",
                        Icon { icon: BsBoxArrowUpRight, width: 18, height: 18 }
                    }
                    div { class: "agents-card-title",
                        {t(l, "developer.agents_api_ref")}
                    }
                    div { class: "agents-card-desc",
                        {t(l, "developer.agents_api_ref_desc")}
                    }
                }
                button { class: "btn-primary", disabled: true, "{t(l, \"developer.launch_agents\")}" }
                span { class: "placeholder-badge", "{t(l, \"common.coming_soon\")}" }
            }
        }
    }
--- a/src/pages/developer/analytics.rs
+++ b/src/pages/developer/analytics.rs
@@ -1,142 +1,40 @@
 use dioxus::prelude::*;
 use dioxus_free_icons::icons::bs_icons::{
    BsBarChartLine, BsBoxArrowUpRight, BsGraphUp, BsSpeedometer,
 };
 use dioxus_free_icons::Icon;
 use crate::i18n::{t, Locale};
-use crate::models::{AnalyticsMetric, ServiceUrlsContext};
+use crate::models::AnalyticsMetric;
-/// Analytics & Observability page for Langfuse.
+/// Analytics page placeholder for LangFuse integration.
 ///
-/// Langfuse is configured with Keycloak SSO (shared realm with CERTifAI).
+/// Shows a "Coming Soon" card with a disabled launch button,
-/// When users open Langfuse, the existing Keycloak session auto-authenticates
+/// plus a mock stats bar showing sample metrics.
 /// them transparently. This page shows a metrics bar, connection status,
 /// and a prominent button to open Langfuse in a new tab.
 #[component]
 pub fn AnalyticsPage() -> Element {
    let locale = use_context::<Signal<Locale>>();
    let svc = use_context::<Signal<ServiceUrlsContext>>();
    let l = *locale.read();
    let url = svc.read().langfuse_url.clone();
    let connected = !url.is_empty();
    let metrics = mock_metrics(l);
    rsx! {
-        div { class: "analytics-page",
+        section { class: "placeholder-page",
            // -- Hero section --
            div { class: "analytics-hero",
                div { class: "analytics-hero-row",
                    div { class: "analytics-hero-icon",
                        Icon { icon: BsGraphUp, width: 24, height: 24 }
                    }
                    h2 { class: "analytics-hero-title",
                        {t(l, "developer.analytics_title")}
                    }
                }
                p { class: "analytics-hero-desc",
                    {t(l, "developer.analytics_desc")}
                }
                // -- Connection status --
                if connected {
                    div { class: "agents-status",
                        span {
                            class: "agents-status-dot agents-status-dot--on",
                        }
                        span { {t(l, "developer.analytics_status_connected")} }
                        code { class: "agents-status-url", {url.clone()} }
                    }
                } else {
                    div { class: "agents-status",
                        span {
                            class: "agents-status-dot agents-status-dot--off",
                        }
                        span { {t(l, "developer.analytics_status_not_connected")} }
                        span { class: "agents-status-hint",
                            {t(l, "developer.analytics_config_hint")}
                        }
                    }
                }
                // -- SSO info --
                if connected {
                    p { class: "analytics-sso-hint",
                        {t(l, "developer.analytics_sso_hint")}
                    }
                }
            }
            // -- Metrics bar --
            div { class: "analytics-stats-bar",
                for metric in &metrics {
                    div { class: "analytics-stat",
                        span { class: "analytics-stat-value", "{metric.value}" }
                        span { class: "analytics-stat-label", "{metric.label}" }
-                        span {
+                        span { class: if metric.change_pct >= 0.0 { "analytics-stat-change analytics-stat-change--up" } else { "analytics-stat-change analytics-stat-change--down" },
                            class: if metric.change_pct >= 0.0 {
                                "analytics-stat-change analytics-stat-change--up"
                            } else {
                                "analytics-stat-change analytics-stat-change--down"
                            },
                            "{metric.change_pct:+.1}%"
                        }
                    }
                }
            }
-
+            div { class: "placeholder-card",
-            // -- Open Langfuse button --
+                div { class: "placeholder-icon", "L" }
-            if connected {
+                h2 { "{t(l, \"developer.analytics_title\")}" }
-                a {
+                p { class: "placeholder-desc",
-                    class: "analytics-launch-btn",
+                    "{t(l, \"developer.analytics_desc\")}"
                    href: "{url}",
                    target: "_blank",
                    rel: "noopener noreferrer",
                    Icon { icon: BsBoxArrowUpRight, width: 16, height: 16 }
                    span { {t(l, "developer.launch_analytics")} }
                }
            }
            // -- Quick actions --
            h3 { class: "agents-section-title",
                {t(l, "developer.analytics_quick_actions")}
            }
            div { class: "agents-grid",
                // Traces
                a {
                    class: if connected { "agents-card" } else { "agents-card agents-card--disabled" },
                    href: if connected { format!("{url}/project") } else { "#".to_string() },
                    target: "_blank",
                    rel: "noopener noreferrer",
                    div { class: "agents-card-icon",
                        Icon { icon: BsBarChartLine, width: 18, height: 18 }
                    }
                    div { class: "agents-card-title",
                        {t(l, "developer.analytics_traces")}
                    }
                    div { class: "agents-card-desc",
                        {t(l, "developer.analytics_traces_desc")}
                    }
                }
                // Dashboard
                a {
                    class: if connected { "agents-card" } else { "agents-card agents-card--disabled" },
                    href: if connected { format!("{url}/project") } else { "#".to_string() },
                    target: "_blank",
                    rel: "noopener noreferrer",
                    div { class: "agents-card-icon",
                        Icon { icon: BsSpeedometer, width: 18, height: 18 }
                    }
                    div { class: "agents-card-title",
                        {t(l, "developer.analytics_dashboard")}
                    }
                    div { class: "agents-card-desc",
                        {t(l, "developer.analytics_dashboard_desc")}
                    }
                }
                button { class: "btn-primary", disabled: true, "{t(l, \"developer.launch_analytics\")}" }
                span { class: "placeholder-badge", "{t(l, \"common.coming_soon\")}" }
            }
        }
    }
--- a/src/pages/developer/flow.rs
+++ b/src/pages/developer/flow.rs
@@ -1,27 +1,27 @@
 use dioxus::prelude::*;
 use crate::components::ToolEmbed;
 use crate::i18n::{t, Locale};
 use crate::models::ServiceUrlsContext;
-/// Flow page embedding the LangFlow visual workflow builder.
+/// Flow page placeholder for the LangFlow visual workflow builder.
 ///
-/// When `langflow_url` is configured, embeds the service in an iframe
+/// Shows a "Coming Soon" card with a disabled launch button.
-/// with a pop-out button. Otherwise shows a "Not Configured" placeholder.
+/// Will eventually integrate with LangFlow for visual flow design.
 #[component]
 pub fn FlowPage() -> Element {
    let locale = use_context::<Signal<Locale>>();
    let svc = use_context::<Signal<ServiceUrlsContext>>();
    let l = *locale.read();
    let url = svc.read().langflow_url.clone();
    rsx! {
-        ToolEmbed {
+        section { class: "placeholder-page",
-            url,
+            div { class: "placeholder-card",
-            title: t(l, "developer.flow_title"),
+                div { class: "placeholder-icon", "F" }
-            description: t(l, "developer.flow_desc"),
+                h2 { "{t(l, \"developer.flow_title\")}" }
-            icon: "F",
+                p { class: "placeholder-desc",
-            launch_label: t(l, "developer.launch_flow"),
+                    "{t(l, \"developer.flow_desc\")}"
                }
                button { class: "btn-primary", disabled: true, "{t(l, \"developer.launch_flow\")}" }
                span { class: "placeholder-badge", "{t(l, \"common.coming_soon\")}" }
            }
        }
    }
 }
--- a/src/pages/organization/dashboard.rs
+++ b/src/pages/organization/dashboard.rs
@@ -2,14 +2,12 @@ use dioxus::prelude::*;
 use crate::components::{MemberRow, PageHeader};
 use crate::i18n::{t, tw, Locale};
-use crate::infrastructure::litellm::get_litellm_usage;
+use crate::models::{BillingUsage, MemberRole, OrgMember};
 use crate::models::{BillingUsage, LitellmUsageStats, MemberRole, OrgMember};
 /// Organization dashboard with billing stats, member table, and invite modal.
 ///
-/// Shows current billing usage (fetched from LiteLLM), a per-model
+/// Shows current billing usage, a table of organization members
-/// breakdown table, a table of organization members with role
+/// with role management, and a button to invite new members.
 /// management, and a button to invite new members.
 #[component]
 pub fn OrgDashboardPage() -> Element {
    let locale = use_context::<Signal<Locale>>();
@@ -22,20 +20,6 @@ pub fn OrgDashboardPage() -> Element {
    let members_list = members.read().clone();
    // Compute date range: 1st of current month to today
    let (start_date, end_date) = current_month_range();
    // Fetch real usage stats from LiteLLM via server function.
    // use_resource memoises and won't re-fire on parent re-renders.
    let usage_resource = use_resource(move || {
        let start = start_date.clone();
        let end = end_date.clone();
        async move { get_litellm_usage(start, end).await }
    });
    // Clone out of Signal to avoid holding the borrow across rsx!
    let usage_snapshot = usage_resource.read().clone();
    // Format token counts for display
    let tokens_display = format_tokens(usage.tokens_used);
    let tokens_limit_display = format_tokens(usage.tokens_limit);
@@ -46,39 +30,26 @@ pub fn OrgDashboardPage() -> Element {
                title: t(l, "org.title"),
                subtitle: t(l, "org.subtitle"),
                actions: rsx! {
-                    button {
+                    button { class: "btn-primary", onclick: move |_| show_invite.set(true), {t(l, "org.invite_member")} }
                        class: "btn-primary",
                        onclick: move |_| show_invite.set(true),
                        {t(l, "org.invite_member")}
                    }
                },
            }
            // Stats bar
            div { class: "org-stats-bar",
                div { class: "org-stat",
-                    span { class: "org-stat-value",
+                    span { class: "org-stat-value", "{usage.seats_used}/{usage.seats_total}" }
                        "{usage.seats_used}/{usage.seats_total}"
                    }
                    span { class: "org-stat-label", {t(l, "org.seats_used")} }
                }
                div { class: "org-stat",
                    span { class: "org-stat-value", "{tokens_display}" }
-                    span { class: "org-stat-label",
+                    span { class: "org-stat-label", {tw(l, "org.of_tokens", &[("limit", &tokens_limit_display)])} }
                        {tw(l, "org.of_tokens", &[("limit", &tokens_limit_display)])}
                    }
                }
                div { class: "org-stat",
-                    span { class: "org-stat-value",
+                    span { class: "org-stat-value", "{usage.billing_cycle_end}" }
                        "{usage.billing_cycle_end}"
                    }
                    span { class: "org-stat-label", {t(l, "org.cycle_ends")} }
                }
            }
            // LiteLLM usage stats section
            {render_usage_section(l, &usage_snapshot)}
            // Members table
            div { class: "org-table-wrapper",
                table { class: "org-table",
@@ -143,144 +114,6 @@ pub fn OrgDashboardPage() -> Element {
    }
 }
 /// Render the LiteLLM usage stats section: totals bar + per-model table.
 ///
 /// Shows a loading state while the resource is pending, an error/empty
 /// message on failure, and the full breakdown on success.
 fn render_usage_section(
    l: Locale,
    snapshot: &Option<Result<LitellmUsageStats, ServerFnError>>,
 ) -> Element {
    match snapshot {
        None => rsx! {
            div { class: "org-usage-loading",
                span { {t(l, "org.loading_usage")} }
            }
        },
        Some(Err(_)) => rsx! {
            div { class: "org-usage-unavailable",
                span { {t(l, "org.usage_unavailable")} }
            }
        },
        Some(Ok(stats)) if stats.total_tokens == 0 && stats.model_breakdown.is_empty() => {
            rsx! {
                div { class: "org-usage-unavailable",
                    span { {t(l, "org.usage_unavailable")} }
                }
            }
        }
        Some(Ok(stats)) => {
            let spend_display = format!("${:.2}", stats.total_spend);
            let total_display = format_tokens(stats.total_tokens);
            // Free-tier LiteLLM doesn't provide prompt/completion split
            let has_token_split =
                stats.total_prompt_tokens > 0 || stats.total_completion_tokens > 0;
            rsx! {
                // Usage totals bar
                div { class: "org-stats-bar",
                    div { class: "org-stat",
                        span { class: "org-stat-value", "{spend_display}" }
                        span { class: "org-stat-label",
                            {t(l, "org.total_spend")}
                        }
                    }
                    div { class: "org-stat",
                        span { class: "org-stat-value",
                            "{total_display}"
                        }
                        span { class: "org-stat-label",
                            {t(l, "org.total_tokens")}
                        }
                    }
                    // Only show prompt/completion split when available
                    if has_token_split {
                        div { class: "org-stat",
                            span { class: "org-stat-value",
                                {format_tokens(stats.total_prompt_tokens)}
                            }
                            span { class: "org-stat-label",
                                {t(l, "org.prompt_tokens")}
                            }
                        }
                        div { class: "org-stat",
                            span { class: "org-stat-value",
                                {format_tokens(stats.total_completion_tokens)}
                            }
                            span { class: "org-stat-label",
                                {t(l, "org.completion_tokens")}
                            }
                        }
                    }
                }
                // Per-model breakdown table
                if !stats.model_breakdown.is_empty() {
                    h3 { class: "org-section-title",
                        {t(l, "org.model_usage")}
                    }
                    div { class: "org-table-wrapper",
                        table { class: "org-table",
                            thead {
                                tr {
                                    th { {t(l, "org.model")} }
                                    th { {t(l, "org.tokens")} }
                                    th { {t(l, "org.spend")} }
                                }
                            }
                            tbody {
                                for model in &stats.model_breakdown {
                                    tr { key: "{model.model}",
                                        td { "{model.model}" }
                                        td {
                                            {format_tokens(model.total_tokens)}
                                        }
                                        td {
                                            {format!(
                                                "${:.2}", model.spend
                                            )}
                                        }
                                    }
                                }
                            }
                        }
                    }
                }
            }
        }
    }
 }
 /// Compute the date range for the current billing month.
 ///
 /// Returns `(start_date, end_date)` as `YYYY-MM-DD` strings where
 /// start_date is the 1st of the current month and end_date is today.
 ///
 /// On the web target this uses `js_sys::Date` to read the browser clock.
 /// On the server target (SSR) it falls back to `chrono::Utc::now()`.
 fn current_month_range() -> (String, String) {
    #[cfg(feature = "web")]
    {
        // js_sys::Date accesses the browser's local clock in WASM.
        let now = js_sys::Date::new_0();
        let year = now.get_full_year();
        // JS months are 0-indexed, so add 1 for calendar month
        let month = now.get_month() + 1;
        let day = now.get_date();
        let start = format!("{year:04}-{month:02}-01");
        let end = format!("{year:04}-{month:02}-{day:02}");
        (start, end)
    }
    #[cfg(not(feature = "web"))]
    {
        use chrono::Datelike;
        let today = chrono::Utc::now().date_naive();
        let start = format!("{:04}-{:02}-01", today.year(), today.month());
        let end = today.format("%Y-%m-%d").to_string();
        (start, end)
    }
 }
 /// Formats a token count into a human-readable string (e.g. "1.2M").
 fn format_tokens(count: u64) -> String {
    const M: u64 = 1_000_000;
--- a/src/pages/providers.rs
+++ b/src/pages/providers.rs
@@ -13,8 +13,8 @@ pub fn ProvidersPage() -> Element {
    let locale = use_context::<Signal<Locale>>();
    let l = *locale.read();
-    let mut selected_provider = use_signal(|| LlmProvider::LiteLlm);
+    let mut selected_provider = use_signal(|| LlmProvider::Ollama);
-    let mut selected_model = use_signal(|| "qwen3-32b".to_string());
+    let mut selected_model = use_signal(|| "llama3.1:8b".to_string());
    let mut selected_embedding = use_signal(|| "nomic-embed-text".to_string());
    let mut api_key = use_signal(String::new);
    let mut saved = use_signal(|| false);
@@ -59,12 +59,12 @@ pub fn ProvidersPage() -> Element {
                                    "Hugging Face" => LlmProvider::HuggingFace,
                                    "OpenAI" => LlmProvider::OpenAi,
                                    "Anthropic" => LlmProvider::Anthropic,
-                                    _ => LlmProvider::LiteLlm,
+                                    _ => LlmProvider::Ollama,
                                };
                                selected_provider.set(prov);
                                saved.set(false);
                            },
-                            option { value: "LiteLLM", "LiteLLM" }
+                            option { value: "Ollama", "Ollama" }
                            option { value: "Hugging Face", "Hugging Face" }
                            option { value: "OpenAI", "OpenAI" }
                            option { value: "Anthropic", "Anthropic" }
@@ -156,28 +156,22 @@ pub fn ProvidersPage() -> Element {
 fn mock_models() -> Vec<ModelEntry> {
    vec![
        ModelEntry {
-            id: "qwen3-32b".into(),
+            id: "llama3.1:8b".into(),
-            name: "Qwen3 32B".into(),
+            name: "Llama 3.1 8B".into(),
-            provider: LlmProvider::LiteLlm,
+            provider: LlmProvider::Ollama,
            context_window: 32,
        },
        ModelEntry {
            id: "llama-3.3-70b".into(),
            name: "Llama 3.3 70B".into(),
            provider: LlmProvider::LiteLlm,
            context_window: 128,
        },
        ModelEntry {
-            id: "mistral-small-24b".into(),
+            id: "llama3.1:70b".into(),
-            name: "Mistral Small 24B".into(),
+            name: "Llama 3.1 70B".into(),
-            provider: LlmProvider::LiteLlm,
+            provider: LlmProvider::Ollama,
-            context_window: 32,
+            context_window: 128,
        },
        ModelEntry {
-            id: "deepseek-r1-70b".into(),
+            id: "mistral:7b".into(),
-            name: "DeepSeek R1 70B".into(),
+            name: "Mistral 7B".into(),
-            provider: LlmProvider::LiteLlm,
+            provider: LlmProvider::Ollama,
-            context_window: 64,
+            context_window: 32,
        },
        ModelEntry {
            id: "meta-llama/Llama-3.1-8B".into(),
@@ -206,7 +200,7 @@ fn mock_embeddings() -> Vec<EmbeddingEntry> {
        EmbeddingEntry {
            id: "nomic-embed-text".into(),
            name: "Nomic Embed Text".into(),
-            provider: LlmProvider::LiteLlm,
+            provider: LlmProvider::Ollama,
            dimensions: 768,
        },
        EmbeddingEntry {
Author	SHA1	Message	Date
Sharang Parnerkar	99967e0082	feat(sidebar): read LibreChat URL from LIBRECHAT_URL env var Some checks failed CI / Clippy (push) Successful in 3m3s Details CI / Security Audit (push) Has been skipped Details CI / Tests (push) Has been skipped Details CI / Format (pull_request) Failing after 3s Details CI / Format (push) Failing after 4s Details CI / Clippy (pull_request) Successful in 3m2s Details CI / Security Audit (pull_request) Has been skipped Details CI / Tests (pull_request) Has been skipped Details CI / Deploy (push) Has been skipped Details CI / Deploy (pull_request) Has been skipped Details Replace hardcoded localhost:3080 chat link with configurable LIBRECHAT_URL environment variable, passed through AuthInfo to the sidebar component. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-24 12:32:57 +01:00
Sharang Parnerkar	833320b21e	feat(librechat): update config URLs for Coolify production deployment Some checks failed CI / Clippy (push) Has been cancelled Details CI / Format (pull_request) Has been cancelled Details CI / Clippy (pull_request) Has been cancelled Details CI / Security Audit (pull_request) Has been cancelled Details CI / Tests (pull_request) Has been cancelled Details CI / Deploy (pull_request) Has been cancelled Details CI / Format (push) Successful in 25s Details CI / Security Audit (push) Has been cancelled Details CI / Tests (push) Has been cancelled Details CI / Deploy (push) Has been cancelled Details Update privacy/terms URLs to dash-dev.meghsakha.com and Ollama baseURL to production endpoint for Coolify deployment. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-24 11:41:41 +01:00
Sharang Parnerkar	79e93eac08	feat(librechat): add OIDC HTTP patch and prompt=none for seamless SSO All checks were successful CI / Format (push) Successful in 2s Details CI / Clippy (push) Successful in 2m45s Details CI / Security Audit (push) Has been skipped Details CI / Tests (push) Has been skipped Details CI / Deploy (push) Has been skipped Details Switch to host networking so LibreChat can reach Keycloak on localhost. Patch openidStrategy.js to allow HTTP OIDC issuers for local dev (openid-client v6 enforces HTTPS by default). Add support for OPENID_AUTH_EXTRA_PARAMS env var and set prompt=none for automatic SSO login when a Keycloak session exists. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-23 22:27:07 +01:00
Sharang Parnerkar	5f3939bb6c	fix(librechat): remove prompt=none for local dev compatibility All checks were successful CI / Format (push) Successful in 2s Details CI / Clippy (push) Successful in 2m47s Details CI / Security Audit (push) Has been skipped Details CI / Tests (push) Has been skipped Details CI / Deploy (push) Has been skipped Details prompt=none causes silent failure when no Keycloak session exists yet. Standard OIDC flow still provides seamless SSO when the user has an active Keycloak session from the dashboard. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-23 21:51:46 +01:00
Sharang Parnerkar	e13afb9d1b	fix(librechat): correct YAML config format and add required secrets Some checks failed CI / Format (push) Successful in 3s Details CI / Security Audit (push) Has been cancelled Details CI / Tests (push) Has been cancelled Details CI / Deploy (push) Has been cancelled Details CI / Clippy (push) Has been cancelled Details - Fix endpoints config: use `custom` array format (not `ollama` key) per LibreChat docs - Add required JWT_SECRET, JWT_REFRESH_SECRET, CREDS_KEY, CREDS_IV environment variables - Update config version to 1.2.8 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-23 21:50:36 +01:00
Sharang Parnerkar	809ca15979	fix(docker): fix keycloak health check and deprecation warnings All checks were successful CI / Format (push) Successful in 3s Details CI / Clippy (push) Successful in 2m56s Details CI / Security Audit (push) Has been skipped Details CI / Tests (push) Has been skipped Details CI / Deploy (push) Has been skipped Details Keycloak 26 image does not include curl, so the health check fails. Switch to bash TCP check against /realms/master. Also replace deprecated KEYCLOAK_ADMIN env vars with KC_BOOTSTRAP_ADMIN_USERNAME/PASSWORD, remove obsolete compose version key, and add start_period for slower machines. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-23 21:47:06 +01:00
Sharang Parnerkar	74a225224c	feat(chat): replace built-in chat with LibreChat SSO integration All checks were successful CI / Format (push) Successful in 26s Details CI / Clippy (push) Successful in 3m0s Details CI / Security Audit (push) Has been skipped Details CI / Tests (push) Has been skipped Details CI / Deploy (push) Has been skipped Details Replaces the custom chat page with an external LibreChat instance that shares Keycloak SSO for seamless auto-login. Removes Tools and Knowledge Base pages as these are now handled by LibreChat's built-in capabilities. - Add LibreChat service to docker-compose with Ollama backend config - Add Keycloak OIDC client (certifai-librechat) with prompt=none for silent SSO - Create librechat.yaml with CERTifAI branding, Ollama endpoint, and custom page title/logo - Change sidebar Chat link to external URL (opens LibreChat in new tab) - Remove chat page, tools page, knowledge base page and all related components (chat_sidebar, chat_bubble, chat_input_bar, etc.) - Remove tool_card, file_row components and tool/knowledge models - Remove chat_stream SSE handler (no longer needed) - Clean up i18n files: remove chat, tools, knowledge sections - Dashboard article summarization via Ollama remains intact Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-23 13:37:17 +01:00