Your agent in your terminal, equipped with local tools: writes code, uses the terminal, browses the web. Make your own persistent autonomous agent on top!
Gptme is an open-source, provider-agnostic AI agent CLI written in Python that runs in any terminal environment, including SSH sessions, tmux, headless servers, and CI pipelines. It connects to Claude (via the Anthropic API, including Claude 3.7 Sonnet) as one of several supported backends alongside OpenAI, Google, DeepSeek, OpenRouter, and local models via llama.cpp. The agent ships with a broad set of built-in tools: shell execution, Python REPL, file read/write, a morph tool for fast targeted edits, web browsing, vision, tmux control, a form tool, background jobs, and auto-commit for Git workflows. It also supports MCP integration for dynamic tool discovery and loading. A plugin system added in v0.30 allows community extensions such as Twitter, Discord, and email tools via the gptme-contrib repository. A companion agent-template project enables persistent autonomous agents, demonstrated by an experimental agent called Bob that monitors GitHub and runs unsupervised loops. Developers, sysadmins, and researchers who want a self-hosted, locally controlled coding and knowledge-work agent without being locked into a single provider or interface are the primary audience.
- ✓Open-source license (MIT)
- ✓Actively maintained (<30d)
- ✓Healthy fork ratio
- ✓Clear description
- ✓Topics declared
- ✓Mature repo (>1y old)
git clone https://github.com/gptme/gptme && cp gptme/*.md ~/.claude/agents/2 items in this repository
Classify codebases before modification to choose appropriate development approach
Subagents overview
What people ask about gptme
What is gptme/gptme?
+
gptme/gptme is subagents for the Claude AI ecosystem. Your agent in your terminal, equipped with local tools: writes code, uses the terminal, browses the web. Make your own persistent autonomous agent on top! It has 4.3k GitHub stars and was last updated today.
How do I install gptme?
+
You can install gptme by cloning the repository (https://github.com/gptme/gptme) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is gptme/gptme safe to use?
+
Our security agent has analyzed gptme/gptme and assigned a Trust Score of 100/100 (tier: Verified). See the full breakdown of passed checks and flags on this page.
Who maintains gptme/gptme?
+
gptme/gptme is maintained by gptme. The last recorded GitHub activity is from today, with 13 open issues.
Are there alternatives to gptme?
+
Yes. On ClaudeWave you can browse similar subagents at /categories/agents, sorted by popularity or recent activity.
Deploy gptme to your cloud
Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.
Maintain this repo? Add a badge to your README
Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.
[](https://claudewave.com/repo/gptme-gptme)<a href="https://claudewave.com/repo/gptme-gptme"><img src="https://claudewave.com/api/badge/gptme-gptme" alt="Featured on ClaudeWave: gptme/gptme" width="320" height="64" /></a>More Subagents
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
The agent that grows with you
Java 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Production-ready platform for agentic workflow development.
The agent engineering platform.
🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.