Find agent skills by outcome

- Name concrete validation: “Run npm test”

“Add unit tests for X”

- Verb-first: “Add…”

“Refactor…”

- If there are unknowns

include a tiny Open questions section (max 3).

<1–3 sentences: what we’re doing

why

- Each checklist item should be a concrete action and

when helpful

- If unsure but not blocked

make a reasonable assumption and proceed.

- Point to likely files/modules: src/...

app/...

- Identify constraints (language

frameworks

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

CONTRIBUTING.md

test('should fail with original bug'

() => {

test('should pass after fix'

() => {

test('should handle edge cases'

() => {

Turn a user prompt into a single

actionable plan delivered in the final assistant message.

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

() => {

OpenAI Create Plan Skill

- System impact (performance

stability

3. Implement minimal

focused fix

- User impact (UX degradation

data loss

For EVERY fixed bug

provide:

- Business impact (compliance

revenue

- [ ] Fix addresses the root cause

not just symptoms

- Security vulnerabilities (SQL injection

XSS

For each identified bug

document:

- Logic errors (incorrect conditions

wrong calculations