Skip to content

kailiu42/coding-agent-instructions

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

title AI coding agent instructions
version 1.0.0
keywords README, instructions, coding-agents

AI Coding Agent Instructions

Languages: English | 中文

This is a rules repository for AI coding agents. Human maintainers can integrate it into their own projects to standardize code style, Git commits, debugging workflows, and related rules.

How to Use

Reference this repository's index.md from the target project's AGENTS.md:

Refers to `/path/to/coding-agent-instructions/index.md` for additional instructions.

Don't read all the instructions at once. Instead read the index.md first, and
refer to the links to each detail instructions only when necessary. Use the
yaml heading for meta information of each markdown file.

Let Your Agent Read This Repository Directly

Clone this repository locally, then point the target project's AGENTS.md to the local clone directory.

Add It to Your Repository

You can create a git submodule that points to this rules repository:

git submodule add https://github.com/kailiu42/coding-agent-instructions .instructions

You can also link it into your repository:

ln -s /path/to/coding-agent-instructions .instructions

When using this approach, point the prompt path to .instructions/index.md.

Except when using git submodules, it is recommended to add .instructions to .gitignore to avoid committing it to your repository.

If You Use My sandboxed-agents

Add this to ~/.config/sandboxed-agents/common.conf:

USER_BINDS+=(
  # Instructions
  --ro-bind-try "/path/to/coding-agent-instructions" "$WORKDIR/.instructions"
)

Then point the prompt path to .instructions/index.md.

It is also recommended to add .instructions to .gitignore to avoid committing it to your repository.

Repository Contents

  • README.md is the human-facing entry point.

  • index.md is the AI agent entry point, containing indexes and links for all domains.

  • AGENTS.md explains how to use an AI agent to manage this repository.

  • Each directory contains instructions for a specific domain.

Rule Organization

Read AGENTS.md to learn how to maintain this rules repository.

Rule Requirements

Read AGENTS.md to learn the requirements and formatting conventions for rules.

About

Rules for coding agents. Modular design, progressive disclosure without eating your context window or distract your agents

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages