Agent skill
acceptance-criteria-verification
Use after implementing features - verifies each acceptance criterion with structured testing and posts verification reports to the GitHub issue
Install this agent skill to your Project
npx add-skill https://github.com/troykelly/codex-skills/tree/main/skills/acceptance-criteria-verification
SKILL.md
Acceptance Criteria Verification
Overview
Systematically verify each acceptance criterion and post structured reports.
Core principle: Every criterion verified. Every verification documented.
Announce at start: "I'm using acceptance-criteria-verification to verify the implementation."
The Verification Process
Step 1: Extract Criteria
Read the issue and extract all acceptance criteria:
# Get issue body
gh issue view [ISSUE_NUMBER] --json body -q '.body'
Parse out criteria (look for - [ ] or - [x] patterns in acceptance criteria section).
Step 2: Plan Verification
For each criterion, determine:
| Criterion | Test Type | How to Verify |
|---|---|---|
| [Criterion 1] | Unit test | Run specific test |
| [Criterion 2] | Integration | API call + response check |
| [Criterion 3] | E2E | Browser automation |
| [Criterion 4] | Manual | Visual inspection |
Step 3: Execute Verification
For each criterion, run the appropriate verification:
Unit/Integration Tests
# Run specific tests
pnpm test --grep "[test pattern]"
# Or run test file
pnpm test path/to/specific.test.ts
E2E Tests
# If using Playwright
npx playwright test [test file]
# If using browser automation MCP
# Use mcp__playwright or mcp__puppeteer
Manual Verification
For criteria requiring visual or interactive verification:
- Start the application
- Navigate to relevant area
- Perform the action
- Capture screenshot if relevant
- Document result
Step 4: Record Results
For each criterion, record:
Criterion: [Text from issue]
Status: PASS | FAIL | PARTIAL | SKIP
Evidence: [Test output, screenshot, observation]
Notes: [Any relevant details]
Step 5: Post Verification Report
Post a structured comment to the issue:
gh issue comment [ISSUE_NUMBER] --body "## Verification Report
**Run**: $(date -u +%Y-%m-%dT%H:%M:%SZ)
**By**: agent
**Commit**: $(git rev-parse --short HEAD)
**Branch**: $(git branch --show-current)
### Results
| # | Criterion | Status | Notes |
|---|-----------|--------|-------|
| 1 | [Criterion text] | PASS | [Notes] |
| 2 | [Criterion text] | FAIL | [What failed] |
| 3 | [Criterion text] | PARTIAL | [What works, what doesn't] |
### Summary
| Status | Count |
|--------|-------|
| PASS | X |
| FAIL | X |
| PARTIAL | X |
| SKIP | X |
| **Total** | **X** |
### Test Output
<details>
<summary>Test Results</summary>
\`\`\`
[test output here]
\`\`\`
</details>
### Next Steps
- [ ] [Action items for failures/partials]
"
Step 6: Update Issue Checkboxes
For each passing criterion, check it off in the issue body:
# Get current body
BODY=$(gh issue view [ISSUE_NUMBER] --json body -q '.body')
# Update checkboxes for passing criteria
# (Implementation depends on body format)
# Update issue
gh issue edit [ISSUE_NUMBER] --body "$NEW_BODY"
Step 7: Update Project Fields
# Update project fields using project-status-sync skill
# Verification status
# - All PASS → Passing
# - Any FAIL → Failing
# - Mix of PASS/PARTIAL → Partial
# Criteria Met count
# - Count of PASS criteria
# Last Verified
# - Current date
# Verified By
# - "agent"
Status Definitions
| Status | Meaning | Action |
|---|---|---|
| PASS | Criterion fully met, verified working | Check off in issue |
| FAIL | Criterion not met, requires fix | Document what failed, return to development |
| PARTIAL | Works with issues, needs improvement | Document issues, may need fix |
| SKIP | Could not verify (blocked, N/A, etc.) | Document reason |
E2E Verification Best Practices
When using browser automation:
- Start fresh - New browser session for each verification
- Capture evidence - Screenshots at key points
- Check visible state - Not just DOM, but visible rendering
- Test error cases - Not just happy path
- Clean up - Close sessions after verification
// Example verification flow (pseudo-code)
await page.goto(appUrl);
await page.click('[data-testid="new-chat"]');
await page.waitForSelector('[data-testid="chat-input"]');
await page.screenshot({ path: 'new-chat-verification.png' });
// Verify expected state
const title = await page.title();
expect(title).toContain('New Chat');
Handling Failures
When criteria fail:
- Document specifically what failed
- Include reproduction steps if not obvious
- Capture error messages or screenshots
- Return to development to fix
- Re-run verification after fix
Do NOT:
- Mark as PASS when it failed
- Skip verification because "it should work"
- Ignore intermittent failures
Verification Checklist
Before completing verification:
- All acceptance criteria evaluated
- Each criterion has clear PASS/FAIL/PARTIAL/SKIP status
- Evidence captured for each (test output, screenshots)
- Verification report posted to issue
- Issue checkboxes updated for passing criteria
- Project fields updated
- If any failures, next steps documented
After Verification
Based on results:
| Overall Result | Next Action |
|---|---|
| All PASS | Proceed to code review |
| Any FAIL | Return to development, fix, re-verify |
| Partial only | Discuss with user - acceptable or needs fix? |
Integration
This skill is called by:
issue-driven-development- Step 8
This skill calls:
project-status-sync- Update verification fieldsissue-lifecycle- Post comments
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
hook-development
Use when the user wants to create Codex workflow hooks (pre/post run gates, tool-use validators, stop checks) or needs guidance on hook scripts and hooks.json configuration.
sentry-setup-ai-monitoring
Setup Sentry AI Agent Monitoring in any project. Use this when asked to add AI monitoring, track LLM calls, monitor AI agents, or instrument OpenAI/Anthropic/Vercel AI/LangChain/Google GenAI. Automatically detects installed AI SDKs and configures the appropriate Sentry integration.
agent-development
Use when the user wants to design Codex agent equivalents (specialized workers/profiles/prompt files), define triggering conditions, or build reusable agent prompts and validation tools.
skill-development
Use when the user wants to create or refine Codex skills, improve skill descriptions, organize skill resources, or follow Codex skill best practices.
sentry-setup-logging
Setup Sentry Logging in any project. Use this when asked to add Sentry logs, enable structured logging, setup console log capture, or integrate logging with Sentry. Supports JavaScript, TypeScript, Python, Ruby, React, Next.js, and other frameworks.
frontend-design
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
Didn't find tool you were looking for?