fire

Delegate implementation tasks to external AI models with preflight and post-execution checks.

68|10|Updated Jul 2, 2026
One-click install
npx skills add https://github.com/tomascupr/sous-chef --skill fire
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: fire
Source: https://github.com/tomascupr/sous-chef/tree/main/skills/fire
Command: npx skills add https://github.com/tomascupr/sous-chef --skill fire

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill solves the bottleneck of manual coding by delegating well-defined implementation tasks to specialized AI workers, allowing the head chef to focus on orchestration and review rather than repetitive coding.

Core Features & Use Cases

  • Automated Delegation: Hands off mechanical tasks like refactors, migrations, and boilerplate generation to specialized models like Codex, Sonnet, or Grok.
  • Verified Execution: Ensures all work is validated against checkable criteria (tests, typechecks) before being accepted into the codebase.
  • Use Case: When you need to perform a large-scale refactor across multiple files, use this Skill to generate a ticket, delegate the implementation to a worker, and automatically review the resulting diff and test output.

Quick Start

Use the fire skill to delegate the implementation of the new authentication module to the Codex worker.

Frequently Asked Questions about fire

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate multi-file refactoring and delegate it to an AI worker?

Multi-file refactoring can be automated by delegating structured implementation tasks to specialized AI models. The system generates a job ticket, hands off the mechanical code migration, and automatically validates the resulting diff against tests before accepting the changes.

How does automated delegation handle code review for AI-generated migrations?

Automated delegation handles code review by enforcing strict oversight through post-execution verification. It validates the AI-generated diff against checkable criteria like typechecks and tests before the code is accepted into the codebase.

Do I need a specific environment to delegate implementation tasks to Codex?

Yes, delegating implementation tasks requires a configured Codex CLI or compatible worker environment. This setup is necessary to manage job directories, generate tickets, and execute diff-based validation for the delegated work.

What is the best way to generate boilerplate code across multiple files without manual coding?

The best way to generate boilerplate code without manual coding is to delegate the task to specialized AI models. This lets you focus on orchestration while the worker handles the mechanical generation and undergoes automated preflight checks.

When should I avoid using AI delegation for code migrations?

You should avoid using AI delegation for code migrations that lack checkable validation criteria. The system requires strict post-execution verification like tests or typechecks to ensure high-fidelity execution before accepting the work.