feat(onboarding): accept modbus:// and opc.tcp:// live-URL refs

The ICS probe targets a device's OT ports (Modbus 502, OPC UA 4840,
EtherNet/IP 44818) via the target's LiveUrl artifact. The wizard's
client-side validation only accepted http(s):// refs, so a PLC endpoint
like modbus://plc:502 was rejected and the probe could never be
onboarded. parse_endpoint already understands the modbus:// scheme;
this just lets the ref through the wizard.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

compliance-agent/src/pipeline/orchestrator.rs

@@ -409,17 +409,11 @@ impl PipelineOrchestrator {
             new_count += self.run_ics_probe(target, &target_id, scan_run_id).await?;
         }
         if plc || ics {
-            // PLC/SPS device: also DAST against a WebVisu / exposed endpoint, but
+            // PLC/SPS device: also DAST against a WebVisu / exposed endpoint. The
-            // only when DAST is actually planned — a device reachable only over an
+            // control-logic scan already consumed the code artifact, so the SAST
-            // industrial protocol (e.g. modbus://) has no web surface to crawl, and
+            // pipeline is not re-run.
-            // running DAST there just fails at reconnaissance. Gating here (not only
-            // at provisioning) also stops a DAST target left over from an earlier
-            // run from re-triggering. The control-logic scan already consumed the
-            // code artifact, so the SAST pipeline is not re-run.
-            if plan.has(ScanType::Dast) {
             self.update_phase(scan_run_id, "dast_scanning").await;
             self.maybe_trigger_dast(&target_id, scan_run_id).await;
-            }
             return Ok(new_count);
         }
 

compliance-core/src/scan_matrix.rs

@@ -14,12 +14,8 @@ use crate::models::{ArtifactKind, OnboardedTarget, ScanType, TargetType};
 pub enum ArtifactRequirement {
     /// Source code — a git repo or a source archive.
     Code,
-    /// A reachable running instance (any live URL / endpoint, scheme-agnostic —
+    /// A reachable running instance (live URL / endpoint).
-    /// e.g. the ICS probe works off the host:port of a modbus:// or http:// ref).
     RunningUrl,
-    /// A reachable **web** endpoint — a live URL with an http(s) scheme. DAST is
-    /// an HTTP crawler, so a modbus:// / opc.tcp:// endpoint does not satisfy it.
-    HttpUrl,
     /// A firmware image / binary blob.
     Firmware,
     /// A PLC project (PLCopen XML or Structured Text).
@@ -138,7 +134,7 @@ fn sast_umbrella() -> Vec<ScanRule> {
 /// The rule set for a target type. Scans that are never applicable to a type are
 /// simply absent (e.g. DAST is not listed for a PLC target).
 pub fn rules_for(target_type: TargetType) -> Vec<ScanRule> {
-    use ArtifactRequirement::{Firmware, HttpUrl, Mobile, Plc, RunningUrl};
+    use ArtifactRequirement::{Firmware, Mobile, Plc, RunningUrl};
     match target_type {
         TargetType::WebApp | TargetType::BackendService => {
             let mut r = sast_umbrella();
@@ -146,7 +142,7 @@ pub fn rules_for(target_type: TargetType) -> Vec<ScanRule> {
                 ScanType::Dast,
                 true,
                 "Dynamic scan of the running endpoint",
-                HttpUrl,
+                RunningUrl,
             ));
             r
         }
@@ -207,7 +203,7 @@ pub fn rules_for(target_type: TargetType) -> Vec<ScanRule> {
                 ScanType::Dast,
                 false,
                 "Dynamic scan of exposed network services (if any)",
-                HttpUrl,
+                RunningUrl,
             ));
             r
         }
@@ -253,7 +249,7 @@ pub fn rules_for(target_type: TargetType) -> Vec<ScanRule> {
                     ScanType::Dast,
                     false,
                     "Dynamic scan of the running device (WebVisu / exposed services)",
-                    HttpUrl,
+                    RunningUrl,
                 ),
                 ScanRule::new(
                     ScanType::IcsProbe,
@@ -289,9 +285,7 @@ pub fn supports_pentest(target_type: TargetType) -> bool {
 fn representative_kind(req: ArtifactRequirement) -> Option<ArtifactKind> {
     match req {
         ArtifactRequirement::Code => Some(ArtifactKind::GitRepo),
-        ArtifactRequirement::RunningUrl | ArtifactRequirement::HttpUrl => {
+        ArtifactRequirement::RunningUrl => Some(ArtifactKind::LiveUrl),
-            Some(ArtifactKind::LiveUrl)
-        }
         ArtifactRequirement::Firmware => Some(ArtifactKind::FirmwareImage),
         ArtifactRequirement::Plc => Some(ArtifactKind::PlcProject),
         ArtifactRequirement::Mobile => Some(ArtifactKind::MobilePackage),
@@ -300,22 +294,11 @@ fn representative_kind(req: ArtifactRequirement) -> Option<ArtifactKind> {
     }
 }
 
-/// Whether a live-URL reference is an http(s) web endpoint (vs. an industrial
-/// endpoint like `modbus://` / `opc.tcp://`, which DAST cannot crawl).
-fn is_http_url(source_ref: &str) -> bool {
-    let s = source_ref.trim();
-    s.starts_with("http://") || s.starts_with("https://")
-}
-
 /// Whether the target carries an artifact that satisfies the requirement.
 fn requirement_satisfied(req: ArtifactRequirement, target: &OnboardedTarget) -> bool {
     match req {
         ArtifactRequirement::Code => target.code_artifact().is_some(),
         ArtifactRequirement::RunningUrl => target.has(ArtifactKind::LiveUrl),
-        ArtifactRequirement::HttpUrl => target
-            .artifacts
-            .iter()
-            .any(|a| a.kind == ArtifactKind::LiveUrl && is_http_url(&a.source_ref)),
         ArtifactRequirement::Firmware => target.has(ArtifactKind::FirmwareImage),
         // A PLC project artifact, or a code artifact (git repo / source archive)
         // holding the control logic as PLCopen XML / ST exports — the common way
@@ -339,10 +322,6 @@ pub fn applicable_scans(target: &OnboardedTarget) -> Vec<ScanOption> {
             let required_artifact = representative_kind(rule.requires);
             let blocked_reason = if satisfied {
                 None
-            } else if rule.requires == ArtifactRequirement::HttpUrl {
-                // A live URL may be present but non-HTTP (e.g. modbus://): be
-                // specific so the user knows DAST needs a web endpoint.
-                Some("no http(s) live URL — DAST needs a web endpoint".to_string())
             } else {
                 Some(match required_artifact {
                     Some(kind) => format!("no {kind} artifact provided"),
@@ -483,49 +462,6 @@ mod tests {
             .is_none());
     }
 
-    #[test]
-    fn plc_with_modbus_url_offers_ics_probe_but_blocks_dast() {
-        // A soft-PLC reachable only over Modbus/TCP (no WebVisu). The ICS probe
-        // is applicable (it works off host:port), but DAST — an HTTP crawler —
-        // must be blocked so it isn't offered/run against a non-web endpoint.
-        let t = target_with(
-            TargetType::PlcSps,
-            vec![Artifact::live_url("modbus://plc-sim:502")],
-        );
-        let opts = applicable_scans(&t);
-        let ics = option(&opts, ScanType::IcsProbe).expect("ics probe offered");
-        assert!(
-            ics.blocked_reason.is_none(),
-            "ICS probe should be unblocked for a modbus:// endpoint"
-        );
-        assert!(!ics.default_on, "ICS probe stays opt-in (default-off)");
-        let dast = option(&opts, ScanType::Dast).expect("dast listed");
-        assert!(
-            dast.blocked_reason.is_some(),
-            "DAST must be blocked without an http(s) endpoint"
-        );
-        assert!(!dast.default_on);
-    }
-
-    #[test]
-    fn plc_with_http_webvisu_offers_both_dast_and_ics_probe() {
-        // A PLC exposing a WebVisu over HTTP: both DAST (web) and the ICS probe
-        // (OT ports on the same host) are applicable.
-        let t = target_with(
-            TargetType::PlcSps,
-            vec![Artifact::live_url("http://plc.local/webvisu")],
-        );
-        let opts = applicable_scans(&t);
-        assert!(option(&opts, ScanType::Dast)
-            .expect("dast offered")
-            .blocked_reason
-            .is_none());
-        assert!(option(&opts, ScanType::IcsProbe)
-            .expect("ics probe offered")
-            .blocked_reason
-            .is_none());
-    }
-
     #[test]
     fn pentest_support_matches_reachable_families() {
         assert!(supports_pentest(TargetType::WebApp));

docs/guide/plc.md

@@ -11,72 +11,6 @@ the control application *and* the device it runs on.
 | A device firmware image | Firmware SBOM / CVE (opt-in) |
 | A reachable endpoint (WebVisu, OPC UA) | DAST / pentest (opt-in) |
 
-## Anatomy: a soft PLC is a SoC + Linux + runtime
-
-A CODESYS controller is **not** a monolithic appliance like a classic Siemens
-S7. It is **PC-based ("soft") control** — commodity silicon running a
-general-purpose Linux, with a **software PLC runtime** as just another process:
-
-| Classic PLC (e.g. Siemens S7) | Soft PLC (CODESYS-on-Yocto, OpenPLC-on-Raspbian) |
-| --- | --- |
-| Proprietary hardware + firmware | Commodity SoC (x86 / ARM) |
-| Proprietary OS | General-purpose Linux (a **Yocto** image, or Raspbian) |
-| Proprietary runtime | Software runtime (**CODESYS Control**, or OpenPLC) |
-| STEP7 / TIA project | IEC 61131-3 control app (ST / LD / FBD / SFC) |
-
-Because of this, the device is built along **two independent tracks**, by
-different people, on different timelines, and shipped separately. It also
-inherits the **entire Linux / IT attack surface on top of** the OT / control
-one — which is exactly why a PLC/SPS target is treated as a **composite**:
-Certifai ingests one artifact per layer and scans each with the right pipeline.
-
-```mermaid
-flowchart TB
-    subgraph TA["Track A · Device platform — built by the hardware OEM / vendor"]
-        direction LR
-        A1["Yocto / OpenEmbedded<br/>BSP + RT kernel"] --> A2["Bake in the CODESYS<br/>Control for Linux runtime"] --> A3["bitbake → device image<br/>.wic / .tar + manifest"]
-    end
-    subgraph TB2["Track B · Control application — built by the machine builder / customer"]
-        direction LR
-        B1["CODESYS IDE<br/>ST / LD / FBD / SFC + WebVisu"] --> B2["Reference CODESYS +<br/>vendor libraries"] --> B3["Compile → download<br/>to device (gateway 11740)"]
-    end
-
-    A3 --> DEV(["Running soft-PLC device<br/>SoC + Linux + runtime + control app<br/>Modbus · OPC UA · EtherNet/IP · WebVisu"])
-    B3 --> DEV
-
-    subgraph CERT["What Certifai scans — one layer per artifact"]
-        direction LR
-        S1["Firmware layer<br/>FirmwareStatic · SBOM · CVE"]
-        S2["Control-logic layer<br/>PLC SAST — ST + FBD/LD"]
-        S3["Control-app SBOM<br/>libraries + runtime → CVE"]
-        S4["Running layer<br/>ICS probe · DAST (WebVisu)"]
-    end
-
-    A3 -. firmware image .-> S1
-    B1 -. PLCopen XML / ST via git .-> S2
-    B2 -. projectarchive (zip) .-> S3
-    DEV -. live URL / provisioned .-> S4
-
-    classDef yocto fill:#fde68a,stroke:#b45309,color:#111
-    classDef codesys fill:#bfdbfe,stroke:#1d4ed8,color:#111
-    classDef dev fill:#e9d5ff,stroke:#7e22ce,color:#111
-    classDef cert fill:#bbf7d0,stroke:#15803d,color:#111
-    class A1,A2,A3 yocto
-    class B1,B2,B3 codesys
-    class DEV dev
-    class S1,S2,S3,S4 cert
-```
-
-::: tip Where Yocto fits
-Yocto is **Track A** — the *build system* for the device platform. It produces
-the Linux image and bakes in the CODESYS runtime, so it is the **firmware
-layer**, entirely separate from the control application. Hand it to Certifai as
-its own **firmware image** artifact (scanned by the firmware pipeline, not the
-PLC pipeline). The device OS need not be Yocto — Raspbian/Debian/Buildroot, or
-even an RTOS / bare-metal, are all possible — but Yocto is the common,
-product-grade industrial choice.
-:::
-
 ## Two ways to deliver the project
 
 You can either **upload** the project when onboarding, or point Certifai at a