feat: add pre-steps and post-steps support to threat detection job

- Add `PostSteps` field to `ThreatDetectionConfig` for post-execution steps
- Change `Steps` field semantics: now runs BEFORE engine execution (pre-steps)
- Update `HasRunnableDetection` to check both `Steps` and `PostSteps`
- Update `parseThreatDetectionConfig` to parse `post-steps` frontmatter field
- Update `buildDetectionJobSteps` to place Steps before and PostSteps after engine execution
- Update JSON schema to add `post-steps` field and update `steps` description
- Update documentation in threat-detection.md and frontmatter-full.md
- Update existing test and add new tests for pre/post step ordering

Agent-Logs-Url: https://github.com/github/gh-aw/sessions/afee0692-364c-4f5e-b0d1-7add5d35cdca

Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>

Author: Copilot

docs/src/content/docs/reference/frontmatter-full.md

@@ -5620,10 +5620,14 @@ safe-outputs:
 
     # Option 2: undefined
 
-    # Array of extra job steps to run after detection
+    # Array of extra job steps to run before engine execution
     # (optional)
     steps: []
 
+    # Array of extra job steps to run after engine execution
+    # (optional)
+    post-steps: []
+
     # Runner specification for the detection job. Overrides agent.runs-on for the
     # detection job only. Defaults to agent.runs-on.
     # (optional)

docs/src/content/docs/reference/threat-detection.md

@@ -77,7 +77,10 @@ safe-outputs:
   threat-detection:
     enabled: true                    # Enable/disable detection
     prompt: "Focus on SQL injection" # Additional analysis instructions
-    steps:                           # Custom detection steps
+    steps:                           # Custom steps run before engine execution
+      - name: Setup Security Gateway
+        run: echo "Connecting to security gateway..."
+    post-steps:                      # Custom steps run after engine execution
       - name: Custom Security Check
         run: echo "Running additional checks"
 ```
@@ -90,7 +93,8 @@ safe-outputs:
 | `prompt` | string | Custom instructions appended to default detection prompt |
 | `engine` | string/object/false | AI engine config (`"copilot"`, full config object, or `false` for no AI) |
 | `runs-on` | string/array/object | Runner for the detection job (default: inherits from workflow `runs-on`) |
-| `steps` | array | Additional GitHub Actions steps to run after AI analysis |
+| `steps` | array | Additional GitHub Actions steps to run **before** AI analysis (pre-steps) |
+| `post-steps` | array | Additional GitHub Actions steps to run **after** AI analysis (post-steps) |
 
 ## AI-Based Detection (Default)
 
@@ -186,13 +190,32 @@ safe-outputs:
 
 ## Custom Detection Steps
 
-Add specialized security scanning tools alongside or instead of AI detection:
+Add specialized security scanning tools alongside or instead of AI detection. You can run steps **before** the AI engine (for setup, gateway connections, etc.) and steps **after** (for additional scanning based on AI results).
+
+### Pre-Steps (`steps:`)
+
+Steps defined under `steps:` run **before** the AI engine executes. Use these for setup tasks such as connecting to a private AI gateway, installing security tools, or preparing artifacts.
 
 ```yaml wrap
 safe-outputs:
   create-pull-request:
   threat-detection:
     steps:
+      - name: Connect to Security Gateway
+        run: |
+          echo "Setting up secure connection to analysis gateway..."
+          # Authentication and connection setup
+```
+
+### Post-Steps (`post-steps:`)
+
+Steps defined under `post-steps:` run **after** the AI engine completes its analysis. Use these for additional security scanning, reporting, or cleanup.
+
+```yaml wrap
+safe-outputs:
+  create-pull-request:
+  threat-detection:
+    post-steps:
       - name: Run Security Scanner
         run: |
           echo "Scanning agent output for threats..."
@@ -206,11 +229,11 @@ safe-outputs:
 
 **Available Artifacts:** Custom steps have access to `/tmp/gh-aw/threat-detection/prompt.txt` (workflow prompt), `agent_output.json` (safe output items), and `aw.patch` (git patch file).
 
-**Execution Order:** Download artifacts → Run AI analysis (if enabled) → Execute custom steps → Upload detection log.
+**Execution Order:** Download artifacts → Execute pre-steps (`steps:`) → Run AI analysis (if enabled) → Execute post-steps (`post-steps:`) → Upload detection log.
 
 ## Example: LlamaGuard Integration
 
-Use Ollama with LlamaGuard 3 for specialized threat detection:
+Use Ollama with LlamaGuard 3 for specialized threat detection running after AI analysis:
 
 ```yaml wrap
 ---
@@ -219,7 +242,7 @@ engine: copilot
 safe-outputs:
   create-pull-request:
   threat-detection:
-    steps:
+    post-steps:
       - name: Ollama LlamaGuard 3 Scan
         uses: actions/github-script@v8
         with:
@@ -261,7 +284,7 @@ safe-outputs:
   threat-detection:
     prompt: "Check for authentication bypass vulnerabilities"
     engine: copilot
-    steps:
+    post-steps:
       - name: Static Analysis
         run: |
           # Run static analysis tool
@@ -273,6 +296,24 @@ safe-outputs:
           path: /tmp/gh-aw/threat-detection/aw.patch
 ```
 
+## Example: Private AI Gateway
+
+Connect to a private AI gateway before running the detection engine:
+
+```yaml wrap
+safe-outputs:
+  create-pull-request:
+  threat-detection:
+    steps:
+      - name: Connect to AI Gateway
+        run: |
+          # Authenticate and set up connection to private AI gateway
+          echo "Setting up gateway connection..."
+          ./scripts/setup-gateway.sh
+    engine:
+      id: copilot
+```
+
 ## Error Handling
 
 **When Threats Are Detected:**

pkg/parser/schemas/main_workflow_schema.json

@@ -7756,7 +7756,14 @@
                 },
                 "steps": {
                   "type": "array",
-                  "description": "Array of extra job steps to run after detection",
+                  "description": "Array of extra job steps to run before engine execution",
+                  "items": {
+                    "$ref": "#/$defs/githubActionsStep"
+                  }
+                },
+                "post-steps": {
+                  "type": "array",
+                  "description": "Array of extra job steps to run after engine execution",
                   "items": {
                     "$ref": "#/$defs/githubActionsStep"
                   }

pkg/workflow/threat_detection.go

@@ -14,7 +14,8 @@ var threatLog = logger.New("workflow:threat_detection")
 // ThreatDetectionConfig holds configuration for threat detection in agent output
 type ThreatDetectionConfig struct {
 	Prompt         string        `yaml:"prompt,omitempty"`        // Additional custom prompt instructions to append
-	Steps          []any         `yaml:"steps,omitempty"`         // Array of extra job steps
+	Steps          []any         `yaml:"steps,omitempty"`         // Array of extra job steps to run before engine execution
+	PostSteps      []any         `yaml:"post-steps,omitempty"`    // Array of extra job steps to run after engine execution
 	EngineConfig   *EngineConfig `yaml:"engine-config,omitempty"` // Extended engine configuration for threat detection
 	EngineDisabled bool          `yaml:"-"`                       // Internal flag: true when engine is explicitly set to false
 	RunsOn         string        `yaml:"runs-on,omitempty"`       // Runner override for the detection job
@@ -24,7 +25,7 @@ type ThreatDetectionConfig struct {
 // that actually executes. Returns false when the engine is disabled and no
 // custom steps are configured, since the job would have nothing to run.
 func (td *ThreatDetectionConfig) HasRunnableDetection() bool {
-	return !td.EngineDisabled || len(td.Steps) > 0
+	return !td.EngineDisabled || len(td.Steps) > 0 || len(td.PostSteps) > 0
 }
 
 // IsDetectionJobEnabled reports whether a detection job should be created for
@@ -108,13 +109,20 @@ func (c *Compiler) parseThreatDetectionConfig(outputMap map[string]any) *ThreatD
 				}
 			}
 
-			// Parse steps field
+			// Parse steps field (pre-execution steps, run before engine execution)
 			if steps, exists := configMap["steps"]; exists {
 				if stepsArray, ok := steps.([]any); ok {
 					threatConfig.Steps = stepsArray
 				}
 			}
 
+			// Parse post-steps field (post-execution steps, run after engine execution)
+			if postSteps, exists := configMap["post-steps"]; exists {
+				if postStepsArray, ok := postSteps.([]any); ok {
+					threatConfig.PostSteps = postStepsArray
+				}
+			}
+
 			// Parse runs-on field
 			if runOn, exists := configMap["runs-on"]; exists {
 				if runOnStr, ok := runOn.(string); ok {
@@ -144,7 +152,7 @@ func (c *Compiler) parseThreatDetectionConfig(outputMap map[string]any) *ThreatD
 				}
 			}
 
-			threatLog.Printf("Threat detection configured with custom prompt: %v, custom steps: %v", threatConfig.Prompt != "", len(threatConfig.Steps) > 0)
+			threatLog.Printf("Threat detection configured with custom prompt: %v, custom pre-steps: %v, custom post-steps: %v", threatConfig.Prompt != "", len(threatConfig.Steps) > 0, len(threatConfig.PostSteps) > 0)
 			return threatConfig
 		}
 	}
@@ -186,21 +194,26 @@ func (c *Compiler) buildDetectionJobSteps(data *WorkflowData) []string {
 	// Step 3: Prepare files - copies agent output files to expected paths
 	steps = append(steps, c.buildPrepareDetectionFilesStep()...)
 
-	// Step 4: Setup threat detection (github-script)
+	// Step 4: Custom pre-steps if configured (run before engine execution)
+	if len(data.SafeOutputs.ThreatDetection.Steps) > 0 {
+		steps = append(steps, c.buildCustomThreatDetectionSteps(data.SafeOutputs.ThreatDetection.Steps)...)
+	}
+
+	// Step 5: Setup threat detection (github-script)
 	steps = append(steps, c.buildThreatDetectionAnalysisStep(data)...)
 
-	// Step 5: Engine execution (AWF, no network)
+	// Step 6: Engine execution (AWF, no network)
 	steps = append(steps, c.buildDetectionEngineExecutionStep(data)...)
 
-	// Step 6: Custom steps if configured
-	if len(data.SafeOutputs.ThreatDetection.Steps) > 0 {
-		steps = append(steps, c.buildCustomThreatDetectionSteps(data.SafeOutputs.ThreatDetection.Steps)...)
+	// Step 7: Custom post-steps if configured (run after engine execution)
+	if len(data.SafeOutputs.ThreatDetection.PostSteps) > 0 {
+		steps = append(steps, c.buildCustomThreatDetectionSteps(data.SafeOutputs.ThreatDetection.PostSteps)...)
 	}
 
-	// Step 7: Upload detection-artifact
+	// Step 8: Upload detection-artifact
 	steps = append(steps, c.buildUploadDetectionLogStep(data)...)
 
-	// Step 8: Parse results, log extensively, and set job conclusion (single JS step)
+	// Step 9: Parse results, log extensively, and set job conclusion (single JS step)
 	steps = append(steps, c.buildDetectionConclusionStep()...)
 
 	threatLog.Printf("Generated %d detection job step lines", len(steps))