Merge pull request #917 from Wikid82/nightly

Weekly: Promote nightly to main (2026-04-06)

Author: Wikid82

.docker/docker-entrypoint.sh

@@ -310,23 +310,23 @@ ACQUIS_EOF
         echo "✗ WARNING: LAPI port configuration may be incorrect"
     fi
 
-    # Update hub index to ensure CrowdSec can start
-    if [ ! -f "/etc/crowdsec/hub/.index.json" ]; then
-        echo "Updating CrowdSec hub index..."
-        timeout 60s cscli hub update 2>/dev/null || echo "⚠️ Hub update timed out or failed, continuing..."
+    # Always refresh hub index on startup (stale index causes hash mismatch errors on collection install)
+    echo "Updating CrowdSec hub index..."
+    if ! timeout 60s cscli hub update 2>&1; then
+        echo "⚠️ Hub index update failed (network issue?). Collections may fail to install."
+        echo "   CrowdSec will still start with whatever index is cached."
     fi
 
     # Ensure local machine is registered (auto-heal for volume/config mismatch)
     # We force registration because we just restored configuration (and likely credentials)
     echo "Registering local machine..."
     cscli machines add -a --force 2>/dev/null || echo "Warning: Machine registration may have failed"
 
-    # Install hub items (parsers, scenarios, collections) if local mode enabled
-    if [ "$SECURITY_CROWDSEC_MODE" = "local" ]; then
-        echo "Installing CrowdSec hub items..."
-        if [ -x /usr/local/bin/install_hub_items.sh ]; then
-            /usr/local/bin/install_hub_items.sh 2>/dev/null || echo "Warning: Some hub items may not have installed"
-        fi
+    # Always ensure required collections are present (idempotent — already-installed items are skipped).
+    # Collections are just config files with zero runtime cost when CrowdSec is disabled.
+    echo "Ensuring CrowdSec hub items are installed..."
+    if [ -x /usr/local/bin/install_hub_items.sh ]; then
+        /usr/local/bin/install_hub_items.sh || echo "⚠️ Some hub items may not have installed. CrowdSec can still start."
     fi
 
     # Fix ownership AFTER cscli commands (they run as root and create root-owned files)

.github/agents/Management.agent.md

@@ -167,23 +167,27 @@ The task is not complete until ALL of the following pass with zero issues:
     - **Base URL**: Uses `PLAYWRIGHT_BASE_URL` or default from `playwright.config.js`
     - All E2E tests must pass before proceeding to unit tests
 
-2. **Local Patch Coverage Preflight (MANDATORY - Before Unit/Coverage Tests)**:
-    - Ensure the local patch report is run first via VS Code task `Test: Local Patch Report` or `bash scripts/local-patch-report.sh`.
-    - Verify both artifacts exist: `test-results/local-patch-report.md` and `test-results/local-patch-report.json`.
-    - Use this report to identify changed files needing coverage before running backend/frontend coverage suites.
-
-3. **Coverage Tests (MANDATORY - Verify Explicitly)**:
+2. **Coverage Tests (MANDATORY - Verify Explicitly)**:
     - **Backend**: Ensure `Backend_Dev` ran VS Code task "Test: Backend with Coverage" or `scripts/go-test-coverage.sh`
     - **Frontend**: Ensure `Frontend_Dev` ran VS Code task "Test: Frontend with Coverage" or `scripts/frontend-test-coverage.sh`
     - **Why**: These are in manual stage of pre-commit for performance. Subagents MUST run them via VS Code tasks or scripts.
     - Minimum coverage: 85% for both backend and frontend.
     - All tests must pass with zero failures.
+    - **Outputs**: `backend/coverage.txt` and `frontend/coverage/lcov.info` — these are required inputs for step 3.
+
+3. **Local Patch Coverage Report (MANDATORY - After Coverage Tests)**:
+    - **Purpose**: Identify uncovered lines in files modified by this task so missing tests are written before declaring Done. This is the bridge between "overall coverage is fine" and "the actual lines I changed are tested."
+    - **Prerequisites**: `backend/coverage.txt` and `frontend/coverage/lcov.info` must exist (generated by step 2). If missing, run coverage tests first.
+    - **Run**: VS Code task `Test: Local Patch Report` or `bash scripts/local-patch-report.sh`.
+    - **Verify artifacts**: Both `test-results/local-patch-report.md` and `test-results/local-patch-report.json` must exist with non-empty results.
+    - **Act on findings**: If patch coverage for any changed file is below **90%**, delegate to the responsible agent (`Backend_Dev` or `Frontend_Dev`) to add targeted tests covering the uncovered lines. Re-run coverage (step 2) and this report until the threshold is met.
+    - **Blocking gate**: 90% overall patch coverage. Do not proceed to pre-commit or security scans until resolved or explicitly waived by the user.
 
 4. **Type Safety (Frontend)**:
     - Ensure `Frontend_Dev` ran VS Code task "Lint: TypeScript Check" or `npm run type-check`
     - **Why**: This check is in manual stage of pre-commit for performance. Subagents MUST run it explicitly.
 
-5. **Pre-commit Hooks**: Ensure `QA_Security` ran `pre-commit run --all-files` (fast hooks only; coverage was verified in step 3)
+5. **Pre-commit Hooks**: Ensure `QA_Security` ran `pre-commit run --all-files` (fast hooks only; coverage was verified in step 2)
 
 6. **Security Scans**: Ensure `QA_Security` ran the following with zero Critical or High severity issues:
    - **Trivy Filesystem Scan**: Fast scan of source code and dependencies

.github/instructions/testing.instructions.md

@@ -12,9 +12,19 @@ instruction files take precedence over agent files and operator documentation.
 
 **MANDATORY**: Before running unit tests, verify the application UI/UX functions correctly end-to-end.
 
-## 0.5 Local Patch Coverage Preflight (Before Unit Tests)
+## 0.5 Local Patch Coverage Report (After Coverage Tests)
 
-**MANDATORY**: After E2E and before backend/frontend unit coverage runs, generate a local patch report so uncovered changed lines are visible early.
+**MANDATORY**: After running backend and frontend coverage tests (which generate
+`backend/coverage.txt` and `frontend/coverage/lcov.info`), run the local patch
+report to identify uncovered lines in changed files.
+
+**Purpose**: Overall coverage can be healthy while the specific lines you changed
+are untested. This step catches that gap. If uncovered lines are found in
+feature code, add targeted tests before completing the task.
+
+**Prerequisites**: Coverage artifacts must exist before running the report:
+- `backend/coverage.txt` — generated by `scripts/go-test-coverage.sh`
+- `frontend/coverage/lcov.info` — generated by `scripts/frontend-test-coverage.sh`
 
 Run one of the following from `/projects/Charon`:
 
@@ -26,11 +36,14 @@ Test: Local Patch Report
 bash scripts/local-patch-report.sh
 ```
 
-Required artifacts:
+Required output artifacts:
 - `test-results/local-patch-report.md`
 - `test-results/local-patch-report.json`
 
-This preflight is advisory for thresholds during rollout, but artifact generation is required in DoD.
+**Action on results**: If patch coverage for any changed file is below 90%, add
+tests targeting the uncovered changed lines. Re-run coverage and this report to
+verify improvement. Artifact generation is required for DoD regardless of
+threshold results.
 
 ### PREREQUISITE: Start E2E Environment
 

.github/renovate.json

@@ -11,6 +11,7 @@
     "development"
 
   ],
+  "postUpdateOptions": ["npmDedupe"],
   "timezone": "America/New_York",
   "dependencyDashboard": true,
   "dependencyDashboardApproval": true,

.github/skills/security-scan-docker-image-scripts/run.sh

@@ -139,7 +139,10 @@ log_info "This may take 30-60 seconds on first run (database download)"
 
 # Run Grype against the SBOM (generated from image, not filesystem)
 # This matches exactly what CI does in supply-chain-pr.yml
+# --config ensures .grype.yaml ignore rules are applied, separating
+# ignored matches from actionable ones in the JSON output
 if grype sbom:sbom.cyclonedx.json \
+    --config .grype.yaml \
     --output json \
     --file grype-results.json; then
     log_success "Vulnerability scan complete"
@@ -149,6 +152,7 @@ fi
 
 # Generate SARIF output for GitHub Security (matches CI)
 grype sbom:sbom.cyclonedx.json \
+    --config .grype.yaml \
     --output sarif \
     --file grype-results.sarif 2>/dev/null || true
 

.github/workflows/auto-add-to-project.yml

@@ -8,6 +8,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.issue.number || github.event.pull_request.number }}
   cancel-in-progress: false
 
+permissions:
+  contents: read
+
 jobs:
   add-to-project:
     runs-on: ubuntu-latest

.github/workflows/auto-changelog.yml

@@ -12,6 +12,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event_name }}-${{ github.event.workflow_run.head_branch || github.head_ref || github.ref_name }}
   cancel-in-progress: true
 
+permissions:
+  contents: write
+
 jobs:
   update-draft:
     runs-on: ubuntu-latest

.github/workflows/auto-label-issues.yml

@@ -8,6 +8,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.issue.number }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   auto-label:
     runs-on: ubuntu-latest

.github/workflows/benchmark.yml

@@ -35,7 +35,7 @@ jobs:
           ref: ${{ github.event.workflow_run.head_sha || github.sha }}
 
       - name: Set up Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6
+        uses: actions/setup-go@4a3601121dd01d1626a1e23e37211e3254c1c06c # v6
         with:
           go-version: ${{ env.GO_VERSION }}
 

.github/workflows/cerberus-integration.yml

@@ -20,6 +20,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.workflow_run.event || github.event_name }}-${{ github.event.workflow_run.head_branch || github.ref }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   cerberus-integration:
     name: Cerberus Security Stack Integration