cook-auto-parallel from smithery.ai

[IMPORTANT] Use TaskCreate to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI MUST ask user whether to skip.

Understand Code First — Search codebase for 3+ similar implementations BEFORE writing any code. Read existing files, validate assumptions with grep evidence, map dependencies via graph trace. Never invent new patterns when existing ones work. MUST READ .claude/skills/shared/understand-code-first-protocol.md for full protocol and checklists.

docs/project-reference/domain-entities-reference.md — Domain entity catalog, relationships, cross-service sync (read when task involves business entities/models) (content auto-injected by hook — check for [Injected: ...] header before reading)
docs/test-specs/ — Test specifications by module (read existing TCs; generate/update test specs via /tdd-spec after implementation)

Plan Quality — Every plan phase MUST include ## Test Specifications section with TC-{FEAT}-{NNN} format. Verify TC satisfaction per phase before marking complete. Plans must include story_points and effort in frontmatter. MUST READ .claude/skills/shared/plan-quality-protocol.md for full protocol and checklists.

Rationalization Prevention — AI consistently skips steps via: "too simple for a plan", "I'll test after", "already searched", "code is self-explanatory". These are EVASIONS — not valid reasons. Plan anyway. Test first. Show grep evidence with file:line. Never combine steps to "save time". MUST READ .claude/skills/shared/rationalization-prevention-protocol.md for full protocol and checklists.

Red Flag STOP Conditions — STOP current approach when: 3+ fix attempts on same issue (root cause not identified), each fix reveals NEW problems (upstream root cause), fix requires 5+ files for "simple" change (wrong abstraction layer), using "should work"/"probably fixed" without verification evidence. After 3 failed attempts, report all outcomes and ask user before attempt #4. MUST READ .claude/skills/shared/red-flag-stop-conditions-protocol.md for full protocol and checklists.

Skill Variant: Variant of /cook — autonomous parallel execution with fullstack-developer subagents.