ReddRadar
Agent skill
cook
Install this agent skill to your Project
npx add-skill https://github.com/diegopacheco/ai-playground/tree/main/pocs/cook-cli-fun/.claude/skills/cook
SKILL.md
Cook — Agent Orchestration CLI
cook wraps your coding agent (Claude Code, Codex, OpenCode) in composable workflows: review loops, repeat passes, parallel races, and task-list orchestration.
Important: Never use --sandbox none. The default sandbox mode (agent) is correct when running as a skill. It preserves the parent agent's security boundaries.
Quick reference
# Single work call
cook "Implement dark mode"
# Review loop (work → review → gate, repeat until DONE)
cook "Implement dark mode" review
# Repeat 3 times
cook "Improve the design" x3
# Race 3 versions, pick the best
cook "Implement dark mode" v3 pick "cleanest implementation"
# Two approaches, pick the winner
cook "Auth with JWT" vs "Auth with sessions" pick "best security"
# Work through a task list
cook "Do the next task in PLAN.md" \
ralph 5 "DONE if all tasks complete, else NEXT"
# Everything composes
cook "Implement dark mode" review v3 "cleanest result"
Operators
Operators compose left to right. Loop operators wrap everything to their left.
Loop operators
| Operator | Effect |
|---|---|
review |
Add a review→gate loop (up to 3 iterations by default) |
review N |
Review loop with up to N iterations |
xN / repeat N |
Run work N times sequentially |
ralph N "gate" |
Outer gate for task-list progression (DONE/NEXT) |
Custom review/gate prompts (positional shorthand):
cook "work prompt" "review prompt" "gate prompt"
cook "work prompt" "review prompt" "gate prompt" "iterate prompt" N
Composition operators
| Operator | Effect |
|---|---|
vN / race N |
N identical runs in parallel worktrees |
vs |
2+ different runs in parallel worktrees |
pick ["criteria"] |
Resolver: pick one winner (default) |
merge ["criteria"] |
Resolver: synthesize all results |
compare |
Resolver: write comparison doc, no merge |
Composition examples
cook "A" vs "B" pick "criteria" # two approaches, pick winner
cook "A" vs "B" merge "best of both" # synthesize both
cook "A" vs "B" compare # comparison doc only
cook "A" v3 "criteria" # race 3, implicit pick
cook "A" x3 vs "B" x3 pick "best" # per-branch loop operators
Flags
--max-iterations N Max review iterations
--work-agent AGENT Per-step agent override
--review-agent AGENT
--work-model MODEL Per-step model override
--review-model MODEL
--hide-request Hide the templated request panel
Prerequisites
Before running cook:
- The project must have
cook initrun (creates COOK.md, .cook/config.json) - For composition operators (vs, vN), the working tree must be clean (commit first)
When to use cook vs doing the work directly
Use cook when:
- The user explicitly asks to "cook" or "let it cook"
- Multiple iterations of refinement are needed (review loops)
- Multiple competing approaches should be tried (races, vs)
- A task list needs sequential progression (ralph)
- The user wants autonomous completion without manual review cycles
Do the work directly when:
- It's a simple, one-shot change
- The user wants to review each step interactively
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
diegopacheco/ai-playground
json-formatter
Validate, format, and minify JSON files when users request JSON validation, formatting, or ask to validate their JSONs
diegopacheco/ai-playground
bruno-generator
Scans the entire codebase, detects all HTTP/API endpoints across Java/Spring Boot, Node/Express, Go/Gin, Rust/Actix+Axum, Python/Django, and generates a complete Bruno API client project with .bru files, sample requests, and environments.
diegopacheco/ai-playground
infra-automation-generator
diegopacheco/ai-playground
leak-detect
Scan code for leaked PII, secrets/credentials, and security vulnerabilities that would get you hacked in production.
diegopacheco/ai-playground
skill-evaluator
This skill should be used when the user asks to "evaluate a skill", "review skill quality", "score my skill", "check skill best practices", "rate my skills", "evaluate all skills", "compare skills", or wants to assess skill quality across criteria like clarity, token efficiency, anti-cheating, quality gates, determinism, scope discipline, error recovery, observability, and idempotency.
diegopacheco/ai-playground
metrics-report
Scan an entire codebase, discover and run all test types, compute hybrid coverage, evaluate quality, and generate a full metrics report website with trends and charts.
Didn't find tool you were looking for?