- Official Resources
- Agent Files & Context
- MCP Servers
- Projects & Tools
- Editor & IDE Integrations
- CLI Usage
- Security & Best Practices
- Community
- Contributing
- Amp - Main Amp website.
- Amp Owner's Manual - Comprehensive official documentation covering installation, usage, and best practices.
- Amp CLI - Official CLI package on npm.
- Amp for VS Code - Official VS Code extension.
- Amp SDK - Official SDK documentation for building on Amp.
- Raising an Agent Podcast - Podcast featuring insights from the Amp team.
- Amp 101 YouTube Playlist - Official video tutorials.
- How to Build an Agent - Guide to building AI agents.
- Amp Chronicle - Official news and announcements.
- Amp Status - Service status page.
- Amp on X - Official Twitter/X account.
AGENTS.md files provide context and instructions to AI coding agents about your codebase. They help agents understand project structure, coding standards, and specific requirements.
Examples of AGENTS.md in open source projects:
- LangGraph - LangChain's graph-based agent framework.
- Zoekt - Fast code search by Sourcegraph.
- Ultimate MCP Client - Comprehensive MCP client implementation.
- MCP Advisor - MCP server advisor tool.
- Use MCP - Official MCP usage examples.
Model Context Protocol (MCP) enables AI agents to interact with external tools and services. These MCP servers extend Amp's capabilities:
- llm-rules MCP - Access Cursor rules dynamically via MCP.
- CleanShot MCP - MCP server for CleanShot X screenshot and recording integration.
- MCP Advisor - Tool to help discover and configure MCP servers.
- Ultimate MCP Client - Comprehensive MCP client for testing and debugging servers.
- Use MCP - Official examples and utilities for MCP usage.
Documentation & Guides:
- Unofficial Amp CLI Documentation - Comprehensive documentation for Amp CLI with examples and best practices.
- File-Based Amp Prompting Workflows - Collection of file-based sub-agent orchestration workflows for Amp.
Development Tools:
- Sandboxed.sh - Self-hosted cloud orchestrator for AI coding agents with isolated Linux environments.
- Unofficial Amp Supervisor - TUI control panel for managing multiple Amp CLI instances in parallel.
- Conductor - Run a bunch of Amps in parallel.
- Amp Code Review CI - Continuous integration tool for automated code reviews using Amp.
- Tokscale - CLI tool for tracking token usage from AmpCode and other coding agents.
- Sniff - Misalignment detection in Vibe Coding loops.
- CodeForge - Golang Development tool built with Amp.
Applications Built with Amp:
- Jazzberry AI - The AI Bug Finder.
- 0.email - AI-native email client that manages your inbox.
- Remote Code - Mobile coding platform that brings AI coding agents to your iPhone.
- SageMap - Interactive belief mapping tool for journal entries.
- VT Chat - Privacy-first AI chat application.
- HTTPie Collection Viewer - Upload and explore your HTTPie collections.
- Quad Ops - Lightweight GitOps framework for podman containers.
- PromptVault - Library for managing prompts and templates in Elixir.
- CircuitPython Deploy - CircuitPython deployment tool.
Browser & Misc:
- Sourcegraph Chrome Extension - Chrome extension for code search and navigation.
- Sourcegraph Amp AUR - Arch Linux AUR package for Sourcegraph Amp.
- amp.nvim - Official Neovim plugin for Amp coding agent.
- nvim-amp - Neovim plugin providing syntax highlighting and support for Amp permission and agent files.
- amp.el - Emacs integration for Amp coding agent.
- Amp ACP - ACP adapter for Amp Code, enabling Amp to work in the Zed editor.
- MyScratchpad VS Code Extension - VS Code extension for global and workspace-specific scratch files.
- Amp Dash X - Raycast extension for organizing and executing Amp Code prompts.
- VibeKanban - CLI tool for managing your Kanban boards.
Amp CLI can be integrated with other command-line tools. Use the -x flag for execute mode or pipe input directly:
# Analyze processes
ps aux > processes.txt | amp -x 'identify processes consuming the most resources in processes.txt'
# Parse whois data
whois example.com | amp -x 'organize and condense this whois information'
# Analyze HTTP headers
curl -Is https://github.com > headers.txt && amp -x 'analyze the http headers in headers.txt and determine the tech stack'
# Check npm dependencies
npm list --json | amp -x 'identify outdated or vulnerable dependencies'When using AI coding agents, consider these security aspects:
- Amp Security Reference - Official security documentation and guidelines.
- Amp Permissions - Control what actions the agent can perform.
Key recommendations:
- Always review AI-generated code before committing
- Never include API keys or secrets in prompts or AGENTS.md files
- Consider running agents in isolated environments for sensitive projects
- Vet MCP servers before installation
- 💬 Relens Community - Join the AI coding agents community to share tips, tricks, and workflows.
Contributions of any kind welcome, just follow the guidelines!
Thanks goes to these contributors!
Originally created and maintained by Justin Dorfman.
