Find agent skills by outcome

- Vague steps (“handle backend”

“do auth”)

- If there are unknowns

include a tiny Open questions section (max 3).

- Verb-first: “Add…”

“Refactor…”

- Each checklist item should be a concrete action and

when helpful

- 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

- Identify constraints (language

frameworks

- If unsure but not blocked

make a reasonable assumption and proceed.

test('should handle edge cases'

() => {

test('should pass after fix'

() => {

test('should fail with original bug'

() => {

For EVERY fixed bug

provide:

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

() => {

- [ ] Fix addresses the root cause

not just symptoms

3. Implement minimal

focused fix

- Performance bottlenecks (N+1 queries

inefficient algorithms)

For each identified bug

document:

- System impact (performance

stability

- Check for existing issue tracking (GitHub Issues

TODO/FIXME/HACK/XXX comments)

- Review linting/formatting configurations (ESLint

Prettier

- Identify testing frameworks (Jest

pytest

- Logic errors (incorrect conditions

wrong calculations

Conduct a thorough analysis of the entire repository to identify

prioritize