AI Tools Cheatsheets

The open-source AI coding handbook for developers: practical cheat sheets, comparisons, prompts, MCP guidance, repo instructions, and engineering workflows for Claude Code, Codex, Cursor, Copilot, Gemini, and modern coding agents.

Compare AI Tools Copy Prompts Browse Workflows Star on GitHub

Claude Code · OpenAI Codex · Cursor · GitHub Copilot · Gemini CLI · MCP · AGENTS.md · CLAUDE.md · Prompts · Workflows

Start With the Most Useful Pages

What you want to do	Best starting page
Pick the right AI coding tool	AI Coding Tools Comparison Matrix
Get productive quickly	Quick Start
Learn Claude Code	Claude Code Cheat Sheet
Learn OpenAI Codex	OpenAI Codex Cheat Sheet
Configure repo instructions	AGENTS.md · CLAUDE.md
Connect tools through MCP	MCP Overview · Useful MCP Servers
Copy reusable prompts	Prompt Library · Master Prompts
Run a real engineering task	Workflow Library
Make a small open-source contribution	Community Hub · Contribute in 10 Minutes

Tool Cheat Sheets

Jump to Claude Code, Codex, Cursor, Copilot, or the full tool index.

Prompts and Workflows

Copy structured prompts and pair them with repeatable review, debugging, testing, and migration workflows.

MCP and Agent Setup

Standardize AGENTS.md, CLAUDE.md, and MCP with least-privilege guidance before enabling broad access.

Popular Engineering Workflows

Code Review

Find bugs, regressions, missing tests, and risky changes before merge.

Bug Fixing

Move from reproduction to root cause, minimal patch, and validation.

Test Generation

Generate targeted tests from behavior, risk, and existing conventions.

Migration and Upgrade

Plan framework and dependency upgrades with checkpoints and rollback thinking.

Choose Your Path

Audience	Start here	Why
Developers	For Developers	Tools, repo instructions, workflows, prompts, and safer next steps.
Teams	For Teams	Standardize tools, prompts, MCP usage, and rollout practices.
Engineering Leaders	For Engineering Leaders	Evaluate adoption, governance, value, and internal handbook strategy.
Platform and Security	For Platform and Security	Review permissions, MCP risk, least privilege, and approval controls.

Safety note: Treat coding agents, MCP servers, shell execution, and repo write access as privileged. Start with Security and Permissions.

Built for Fast-Moving AI Tooling

This handbook uses explicit verification labels because AI coding tools change quickly. Verified, Locally tested, Documentation verified, Needs verification, Experimental, account/plan constraints, and similar labels communicate how much confidence to place in a page before standardizing on it.

When a page includes a Sources section, treat those links as the evidence boundary for its claims.

Fork It for Your Team

Use the public handbook as a neutral baseline, then customize approved tools, repo instructions, MCP servers, prompts, workflows, and policies for your organization.

Contribute

A first contribution can be one corrected command, official source, validation step, prompt improvement, accessibility fix, or clearer example.

The public issue queue is intentionally curated: starter issues should name the file, acceptance criteria, source expectations, and validation command. If the handbook saves you time, star the repository so you can find it again and more developers can discover it.