GitHub Actions
28865a5f36
Forced workflow failure if scan results are missing (prevents false negatives) Fixed "Fail on critical" step to use calculated counts instead of missing action outputs Added debug logging and file verification for Grype scans Refactored shell scripts to prevent injection vulnerabilities
245 lines
10 KiB
Markdown
245 lines
10 KiB
Markdown
# QA Report - Supply Chain Workflow Audit
|
|
|
|
**Date:** February 6, 2026
|
|
**Target:** `.github/workflows/supply-chain-pr.yml`
|
|
**Trigger:** Manual Lint Request
|
|
**Auditor:** QA Security Engineer (Gemini 3 Pro)
|
|
|
|
## 1. Executive Summary
|
|
|
|
A manual audit and linting session was performed on the `supply-chain-pr.yml` workflow. Critical logic errors were identified that would have prevented the workflow from correctly downloading artifacts during a PR event. Security vulnerabilities related to script injection were also mitigated.
|
|
|
|
**Status:** 🟡 **REMEDIATED** (Issues found and fixed)
|
|
|
|
## 2. Findings & Remediation
|
|
|
|
### A. Logic Error: Circular Dependency
|
|
* **Severity:** 🔴 **CRITICAL**
|
|
* **Issue:** The steps "Download PR image artifact" and "Load Docker image" conditionally depended on `steps.set-target.outputs.image_name`. However, the `set-target` step is defined **after** these steps in the workflow execution order.
|
|
* **Impact:** These steps would invariably evaluate to `false` or crash, causing the workflow to skip image verification for PRs.
|
|
* **Fix:** Updated the conditions to depend on `steps.check-artifact.outputs.artifact_found == 'true'`, which is correctly populated by the preceding step.
|
|
|
|
### B. Security: Script Injection Risk
|
|
* **Severity:** 🟠 **HIGH**
|
|
* **Issue:** User-controlled inputs (`github.head_ref`, `inputs.pr_number`) were used directly in inline scripts (`run` blocks).
|
|
* **Impact:** A malicious branch name or PR number could potentially execute arbitrary commands in the runner environment.
|
|
* **Fix:** Mapped all user inputs to environment variables (`env` block) and referenced them via shell variables (e.g., `${BRANCH_NAME}`) instead of template injection.
|
|
|
|
### C. Syntax & Linting
|
|
* **Tool:** `actionlint`
|
|
* **Result:** Identified the logic errors and security warnings mentioned above.
|
|
* **Status:** All reported errors logic/security errors addressed. Shellcheck style warnings (redirects) noted but lower priority.
|
|
|
|
### D. Security Scan (Trivy)
|
|
* **Tool:** `trivy fs`
|
|
* **Command:** `trivy fs --scanners secret,misconfig .github/workflows/supply-chain-pr.yml`
|
|
* **Result:** ✅ **PASS**
|
|
* No secrets detected.
|
|
* No infrastructure misconfigurations detected by Trivy policies.
|
|
|
|
## 3. Verification
|
|
The workflow file has been updated with the fixes. It is recommended to trigger a test run (via PR or workflow_dispatch) to verify the runtime behavior.
|
|
|
|
---
|
|
|
|
# QA Report - Phase 6 Audit (Playwright Config Update)
|
|
|
|
**Date:** February 6, 2026
|
|
**Trigger:** Update of `playwright.config.js` to separate and sequence security tests.
|
|
**Auditor:** QA Security Engineer (Gemini 3 Pro)
|
|
|
|
## 1. Executive Summary
|
|
|
|
The Phase 6 Audit was performed to validate the new Playwright configuration which splits security tests into a separate project that runs prior to standard browser tests.
|
|
|
|
**Status:** 🔴 **FAILED**
|
|
|
|
While the configuration successfully enforced the execution order (security tests ran first), the security tests themselves failed due to authentication issues in the test environment. This failure, combined with the new dependency structure, caused the majority of the standard E2E suite (1964 tests) to be skipped.
|
|
|
|
Security scans identified 1 High-severity misconfiguration in the Dockerfile and 2 High-severity vulnerabilities in the container base image.
|
|
|
|
## 2. E2E Test Execution Analysis
|
|
|
|
### Execution Order Verification
|
|
* **Result:** ✅ **Verified**
|
|
* **Observation:** The `security-tests` project executed before `chromium`, `firefox`, and `webkit` projects as configured.
|
|
|
|
### Test Results
|
|
* **Total Tests Run:** 219
|
|
* **Passed:** 201
|
|
* **Failed:** 18
|
|
* **Skipped / Not Run:** 1,964
|
|
* **Pass Rate:** ~9% (of total suite) / 91% (of executed tests)
|
|
|
|
### Failure Analysis
|
|
The 18 failed tests were all within the `security-tests` project. The failures were consistent `401 Unauthorized` errors during test setup/teardown helpers.
|
|
|
|
**Key Error:**
|
|
```
|
|
Failed to enable Cerberus: Error: Failed to set cerberus to true: 401 {"error":"Authorization header required"}
|
|
```
|
|
|
|
**Impacted Areas:**
|
|
1. **Security Helpers:** `setSecurityModuleEnabled()`, `getSecurityStatus()`, `configureAdminWhitelist()` in `tests/utils/security-helpers.ts`.
|
|
2. **Tests:**
|
|
* `security-enforcement/acl-enforcement.spec.ts`
|
|
* `security-enforcement/combined-enforcement.spec.ts`
|
|
* `security-enforcement/crowdsec-enforcement.spec.ts`
|
|
* `security-enforcement/rate-limit-enforcement.spec.ts`
|
|
* `security-enforcement/waf-enforcement.spec.ts`
|
|
* `security/acl-integration.spec.ts` (Also failed finding UI modals)
|
|
|
|
**Root Cause Hypothesis:**
|
|
The test environment (`charon-e2e` container) requires authentication for the management API (`/api/v1/security/*`), but the test helper functions are failing to provide a valid Authorization header or session cookie in the current context.
|
|
|
|
**Blocking Issue:**
|
|
Because `chromium` etc. depend on `security-tests`, the failure of the security suite prevented the standard browser tests from running.
|
|
|
|
## 3. Security Scan Findings
|
|
|
|
### Trivy Filesystem Scan
|
|
* **Command:** `trivy fs /projects/Charon --skip-dirs .cache`
|
|
* **Findings:**
|
|
* **Dockerfile:** 1 🔴 HIGH Misconfiguration
|
|
* **ID:** DS-0002
|
|
* **Message:** "Image user should not be 'root'"
|
|
* **Resolution:** Add `USER <non-root>` instruction.
|
|
|
|
### Trivy Docker Image Scan
|
|
* **Target:** `charon:local` (Debian 13.3)
|
|
* **Findings:**
|
|
* **Total:** 2 🔴 HIGH Vulnerabilities
|
|
* **CVE-2026-0861** (`libc-bin`, `libc6`): Integer overflow in `memalign` leading to heap corruption.
|
|
* **Status:** Fix available in upstream Debian (upgrade required).
|
|
|
|
## 4. Recommendations & Next Steps
|
|
|
|
### Immediate Actions (Blockers)
|
|
1. **Fix Test Authentication:** Investigate `tests/utils/security-helpers.ts`. Ensure it properly authenticates (e.g., logs in via UI or uses a valid API token) before attempting to configure security modules. Inspect `.env` usage in the E2E container.
|
|
2. **Fix UI Interaction:** Investigate `waitForModal` failures in `acl-integration.spec.ts`. The UI might have changed, breaking the locator `"/edit|proxy/i"`.
|
|
|
|
### Security Remediation
|
|
1. **Dockerfile Hardening:** implementation of a non-root user in the `Dockerfile`.
|
|
2. **Base Image Update:** Re-pull the base image (`debian:bookworm-slim` or equivalent) to pick up the patch for CVE-2026-0861, or ensure `apt-get upgrade` runs during build.
|
|
|
|
### Configuration Adjustment
|
|
* **Consider Fail-Open for Dev:** While serial execution is good for CI, consider if local development requires `dependencies: ['security-tests']` to be strict, or if we can allow specific headers/tokens to bypass this for easier debugging.
|
|
|
|
## 5. Conclusion
|
|
The separation of security tests is sound, but the current state of the security test suite is unstable. Prioritize fixing the 401 errors in the security helpers to unblock the rest of the E2E suite.
|
|
|
|
---
|
|
|
|
# QA Report: Project Health Check (Previous)
|
|
|
|
**Date**: 2026-02-05
|
|
**Version**: v0.18.13
|
|
**Scope**: Full project health check via pre-commit hooks and YAML validation.
|
|
|
|
---
|
|
|
|
## Executive Summary
|
|
|
|
| Category | Status | Details |
|
|
|----------|--------|---------|
|
|
| YAML Syntax | ✅ PASS | All YAML files are valid |
|
|
| Pre-commit Hooks | ✅ PASS | All hooks passed (after version fix) |
|
|
| Version Sync | ✅ PASS | `.version` synced with git tag `v0.18.13` |
|
|
| File Consistency | ✅ PASS | No trailing whitespace or end-of-file issues |
|
|
| LFS Usage | ✅ PASS | No untracked large files |
|
|
|
|
**Overall Status**: ✅ **APPROVED** - The codebase is clean and compliant with all quality gates.
|
|
|
|
---
|
|
|
|
## 1. YAML Syntax Validation
|
|
|
|
### Results
|
|
- **Status**: ✅ PASS
|
|
- **Command**: `pre-commit run check-yaml --all-files`
|
|
- **Output**:
|
|
```
|
|
check yaml...............................................................Passed
|
|
```
|
|
|
|
### Analysis
|
|
- All YAML files (workflows, config, docker-compose) are syntactically correct.
|
|
- No parsing errors detected.
|
|
|
|
---
|
|
|
|
## 2. Pre-commit Hook Validation
|
|
|
|
### Results
|
|
- **Status**: ✅ PASS
|
|
- **Command**: `pre-commit run --all-files` (alias `qa-precommit-all`)
|
|
- **Issues Found**:
|
|
- **Initial Run**: ❌ FAIL - `.version` (v0.17.1) did not match Git tag (v0.18.13).
|
|
- **Resolution**: Updated `.version` file to `v0.18.13`.
|
|
- **Final Run**: ✅ PASS
|
|
|
|
### Hook Details
|
|
|
|
| Hook | Status | Notes |
|
|
|------|--------|-------|
|
|
| fix end of files | ✅ Pass | |
|
|
| trim trailing whitespace | ✅ Pass | |
|
|
| check yaml | ✅ Pass | |
|
|
| check for added large files | ✅ Pass | |
|
|
| dockerfile validation | ✅ Pass | |
|
|
| Go Vet | ✅ Pass | |
|
|
| golangci-lint (Fast) | ✅ Pass | |
|
|
| Check .version matches tag | ✅ Pass | Fixed: synced to v0.18.13 |
|
|
| LFS large files check | ✅ Pass | |
|
|
| Prevent CodeQL DB commits | ✅ Pass | |
|
|
| Prevent data/backups commits | ✅ Pass | |
|
|
| Frontend TypeScript Check | ✅ Pass | |
|
|
| Frontend Lint (Fix) | ✅ Pass | |
|
|
|
|
---
|
|
|
|
## 3. Version Synchronization
|
|
|
|
### Issue Detected
|
|
The `.version` file contained `v0.17.1` while the latest git tag was `v0.18.13`, causing the version check hook to fail.
|
|
|
|
### Remediation
|
|
Executed:
|
|
```bash
|
|
echo "v0.18.13" > /projects/Charon/.version
|
|
```
|
|
This aligns the project version file with the source control tag.
|
|
|
|
---
|
|
|
|
## 4. Final Verification
|
|
|
|
A final run of all checks confirmed the project is in a consistent state:
|
|
|
|
```
|
|
fix end of files.........................................................Passed
|
|
trim trailing whitespace.................................................Passed
|
|
check yaml...............................................................Passed
|
|
check for added large files..............................................Passed
|
|
dockerfile validation....................................................Passed
|
|
Go Vet...................................................................Passed
|
|
golangci-lint (Fast Linters - BLOCKING)..................................Passed
|
|
Check .version matches latest Git tag....................................Passed
|
|
Prevent large files that are not tracked by LFS..........................Passed
|
|
Prevent committing CodeQL DB artifacts...................................Passed
|
|
Prevent committing data/backups files....................................Passed
|
|
Frontend TypeScript Check................................................Passed
|
|
Frontend Lint (Fix)......................................................Passed
|
|
```
|
|
|
|
## 5. Recommendations
|
|
|
|
1. **Commit Changes**: Commit the updated `.version` file.
|
|
2. **Proceed**: The codebase is ready for further development or release processes.
|
|
|
|
---
|
|
|
|
*QA Report generated: 2026-02-05*
|
|
*Agent: QA Security Engineer*
|
|
*Validation Type: Health Check*
|