.codex
Defines 9 coding rules and a git workflow for an AI code assistant (Codex) in a Rust monorepo.
What this file does
Defines 9 coding rules and a git workflow for an AI code assistant (Codex) in a Rust monorepo.
When to use it
- You use an AI code assistant that can stage new files
- You want to enforce no global state and no filesystem access in core crates
- You need a lightweight contract for AI to avoid feature creep
- You want to prevent AI from committing or rewriting git history
Assumes this stack
Rules for this repository:
- Never add new crates without being asked.
- Never bypass existing module boundaries.
- Prefer small, testable functions.
- All parsing code must have tests.
- No filesystem access in stbl_core.
- No global state.
- No "TODO" placeholders.
- No feature creep beyond the prompt.
- Always tests run and make sure that the tests pass.
Git workflow (important)
- Codex MAY run
git addfor new files it creates. - Codex MUST NOT run
git addon modified existing files. - Codex MUST NOT create commits (
git commit). - Codex MUST NOT rewrite history.
- All commits are done manually after review.
Rationale:
- New files are easy to miss; adding them helps review.
- Modified files must always be reviewed before staging.
Specifications FYI
AGENTS.md ARCHITECTURE.md specs/*
What's inside
9 rules, 4 git workflow directives, 1 rationale paragraph, 3 specification file references
Change this for your project
- Replace
stbl_corewith your own core crate name - Replace
jgaa/stblrepository references with your own repo URL - Replace
AGENTS.md,ARCHITECTURE.md,specs/*with your own spec file paths
Where it goes
Keep in docs/ or alongside the feature. Agents read it to implement against a defined contract.
Related Documents
Community AI Agent Skills Discovery Sources
Catalogs 50+ platforms, repositories, directories, and communities for discovering and sharing AI agent skills across multiple coding tools.
GPU Selection Guide for Large Language Models (LLMs)
Guides GPU selection for LLM inference, fine-tuning, and training by mapping model sizes, precision levels, and budgets to VRAM requirements.
Python Development Environment Setup with Ruff
Defines a complete Python development environment using Ruff for linting and formatting, with pytest, mypy, pre-commit hooks, and VSCode configuration.
ReleaseKit - Technical Requirements Document
Specifies a Go library and CLI for release automation with conventional commit parsing, validation checks, and workflow orchestration.