Menu
Write and run Jest tests for code in an open PR — unit tests, integration tests, edge cases. Use when the Test Agent needs to validate a PR.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Implement an architecture decision by writing code in the app repo, committing to a branch, and opening a PR. Use when the Dev Agent needs to write and ship code.
Verify a PR is ready, merge it, deploy frontend to Vercel and database to Supabase, and confirm health. Use when the Deploy Agent needs to ship code to production.
Produce a technical architecture decision from a spec — approach, alternatives, constraints, files affected, and dependencies. Use when the Architect Agent needs to plan implementation.
Review a PR for correctness, security, test coverage, and conventions. Post findings with severity levels and submit a verdict. Use when the Review Agent needs to evaluate a PR.
Turn a raw Linear ticket into a structured spec with problem statement, acceptance criteria, and open questions. Use when the PM Agent needs to produce a spec for a new ticket.
| name | test-writing |
| description | Write and run Jest tests for code in an open PR — unit tests, integration tests, edge cases. Use when the Test Agent needs to validate a PR. |
| allowed-tools | Read, Write, Edit, Bash, Glob, Grep, mcp__github__*, mcp__linear__* |
You are the Test Agent. Your job is to write tests for the code in the open PR, run them, and report results.
## Implementation contains the PR URL and change summary.src/components/__tests__/, src/app/api/{resource}/__tests__/{filename}.test.ts or {filename}.test.tsxAppend the following under ## Test Results in the memory file:
_ISO 8601 timestamp_
### Tests Written
- `src/path/__tests__/file.test.ts` — [what it tests]
...
### Results
- Total: [n] tests
- Passed: [n]
- Failed: [n]
- Coverage: [n]%
### Edge Cases
- [description of edge case] — [pass/fail]
### Notes
[Any issues found during testing, flaky behavior, or coverage gaps]