Files
breakpilot-compliance/backend-compliance/tests/test_specialist_agent_routes.py
T
Benjamin Admin 3ae4e60c9d
CI / detect-changes (push) Successful in 7s
CI / branch-name (push) Has been skipped
CI / guardrail-integrity (push) Has been skipped
CI / secret-scan (push) Has been skipped
CI / dep-audit (push) Has been skipped
CI / sbom-scan (push) Has been skipped
CI / build-sha-integrity (push) Failing after 4s
CI / validate-canonical-controls (push) Successful in 12s
CI / loc-budget (push) Successful in 14s
CI / go-lint (push) Has been skipped
CI / python-lint (push) Has been skipped
CI / nodejs-lint (push) Has been skipped
CI / nodejs-build (push) Successful in 2m24s
CI / test-go (push) Has been skipped
CI / iace-gt-coverage (push) Has been skipped
CI / test-python-backend (push) Successful in 29s
CI / test-python-document-crawler (push) Has been skipped
CI / test-python-dsms-gateway (push) Has been skipped
feat(agents): SSE-Endpoint + Agent-Test-Tab (5-URL parallel)
Backend:
- specialist_agent_routes.py: GET /agents, POST /test/start (run_id),
  GET /test/stream/{run_id} (SSE), GET /run/{run_id}/result,
  GET /run/{run_id}/artifacts, GET /run/{run_id}/artifact/{path},
  DELETE /run/{run_id}, GET /runs.
- Per-URL async orchestrator: text fetch via consent-tester
  dsi-discovery → agent.evaluate() → vault.put_json + stream events.
- Tests: 7/7 grün.

Frontend:
- /api/sdk/v1/specialist-agent proxy mit SSE-passthrough.
- AgentTestTab.tsx: Agent-Wähler + 5 URL-Slots + Live-Events +
  Speedometer (OK/N-A/HIGH/MEDIUM/LOW) + Findings + Recommendations +
  Eskalations-Log + Artefakt-Link pro Slot.
- Neuer Tab "Agent-Test" in /sdk/agent.

User-Wunsch 2026-06-08: pro Agent isoliert testen, 5 URLs gleichzeitig,
Live-Updates statt Polling-Wartespiel.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-06-08 17:47:05 +02:00

130 lines
4.3 KiB
Python

"""Tests für SSE-Endpoints des Specialist-Agent-Test-Harness."""
from __future__ import annotations
import asyncio
import json
from unittest.mock import AsyncMock, patch
import pytest
from fastapi.testclient import TestClient
@pytest.fixture
def app(tmp_path, monkeypatch):
monkeypatch.setenv("EVIDENCE_VAULT_ROOT", str(tmp_path / "vault"))
from fastapi import FastAPI
from compliance.api.specialist_agent_routes import router
app = FastAPI()
app.include_router(router, prefix="/api/v1")
return app
@pytest.fixture
def client(app):
return TestClient(app)
def test_list_agents(client):
r = client.get("/api/v1/specialist-agent/agents")
assert r.status_code == 200
data = r.json()
agent_ids = {a["agent_id"] for a in data["agents"]}
assert "impressum" in agent_ids
assert "cookie_policy" in agent_ids
def test_start_test_invalid_agent(client):
r = client.post("/api/v1/specialist-agent/test/start",
json={"agent_id": "ghost",
"raw_texts": ["test"]})
assert r.status_code == 404
def test_start_test_no_input(client):
r = client.post("/api/v1/specialist-agent/test/start",
json={"agent_id": "impressum"})
assert r.status_code == 400
def test_start_test_with_raw_text(client):
r = client.post("/api/v1/specialist-agent/test/start",
json={"agent_id": "impressum",
"raw_texts": ["Tesla Germany GmbH "
"Berlin Email: x@y.com "
"HRB 123 Charlottenburg"]})
assert r.status_code == 200
data = r.json()
assert data["agent_id"] == "impressum"
assert data["slot_count"] == 1
assert data["run_id"]
def test_stream_unknown_run(client):
r = client.get("/api/v1/specialist-agent/test/stream/ghost")
assert r.status_code == 404
def test_run_result_after_text_input(client, monkeypatch):
# Skip LLM
async def _no_cascade(*a, **kw): return None, []
monkeypatch.setattr(
"compliance.services.specialist_agents.impressum.agent.cascade",
_no_cascade,
)
r = client.post("/api/v1/specialist-agent/test/start",
json={"agent_id": "impressum",
"raw_texts": [
"Tesla Germany GmbH\nLudwig-Prandtl-Strasse 25\n"
"12526 Berlin\nDeutschland\nEmail: x@y.com\n"
"Tel: +49 89 1250 16 800\n"
"Management: Elon Musk\n"
"HRB 218904 B Charlottenburg",
]})
run_id = r.json()["run_id"]
# Give async task time to finish (small text → fast)
for _ in range(40):
rr = client.get(
f"/api/v1/specialist-agent/run/{run_id}/result",
)
if rr.json().get("finished"):
break
import time; time.sleep(0.05)
res = client.get(f"/api/v1/specialist-agent/run/{run_id}/result")
body = res.json()
assert body["finished"]
assert "text1" in body["results"]
out = body["results"]["text1"]
field_ids = {f["field_id"] for f in out["findings"]}
# Tesla pattern: German-label fehlt + USt fehlt
assert "vertretungsberechtigte_label_korrekt" in field_ids
def test_artifacts_listing(client, monkeypatch):
async def _no_cascade(*a, **kw): return None, []
monkeypatch.setattr(
"compliance.services.specialist_agents.impressum.agent.cascade",
_no_cascade,
)
r = client.post("/api/v1/specialist-agent/test/start",
json={"agent_id": "impressum",
"raw_texts": ["Tesla Germany GmbH "
"Berlin Email: x@y.com "
"HRB 123 Charlottenburg"]})
run_id = r.json()["run_id"]
for _ in range(40):
rr = client.get(
f"/api/v1/specialist-agent/run/{run_id}/result",
)
if rr.json().get("finished"):
break
import time; time.sleep(0.05)
arts = client.get(
f"/api/v1/specialist-agent/run/{run_id}/artifacts",
)
assert arts.status_code == 200
manifest = arts.json()["manifest"]
kinds = {a["kind"] for a in manifest["assets"]}
assert "finding" in kinds
assert "raw" in kinds