Find agent skills by outcome

- Read README.md and any obvious docs (docs/

CONTRIBUTING.md

Turn a user prompt into a single

actionable plan delivered in the final assistant message.

OpenAI Create Plan Skill

test('should handle edge cases'

() => {

describe('BUG-[ID]: [Bug description]'

() => {

test('should pass after fix'

() => {

test('should fail with original bug'

() => {

3. Implement minimal

focused fix

- System impact (performance

stability

- User impact (UX degradation

data loss

- [ ] Fix addresses the root cause

not just symptoms

- Business impact (compliance

revenue

For EVERY fixed bug

provide:

For each identified bug

document:

- State management issues (race conditions

inconsistent state

- Performance bottlenecks (N+1 queries

inefficient algorithms)

- Security vulnerabilities (SQL injection

XSS

- Logic errors (incorrect conditions

wrong calculations

- Review linting/formatting configurations (ESLint

Prettier

- Check for existing issue tracking (GitHub Issues

TODO/FIXME/HACK/XXX comments)

- Identify testing frameworks (Jest

pytest

- Identify technology stack and dependencies (package.json

requirements.txt

- Map complete project structure (src/

lib/

- Document main entry points

critical paths