BigHugger
GH Repository · gptme

gptme

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!

stars
4,417
30-day movement
+10/day
Related entries
60
Connections
2
agent-appPythonagentschatbotpythonllmopenrouteranthropicpython/poetryllamacppopenaimakedockerai-agentschatgptcode-generationllm-appscliragagentai-assistantllm-agent

gptme is a terminal-based CLI agent that runs LLM-powered assistants with local tools for writing code, using the shell, and browsing the web. It supports multiple providers (OpenAI, Anthropic, OpenRouter, llama.cpp) and is designed as a base for building your own persistent autonomous agent.

You want an LLM agent that works directly in your terminal with local tool access and provider flexibility.

Use it to

  • Run a coding assistant in the terminal
  • Let an agent execute shell commands
  • Build a custom persistent autonomous agent
  • Chat with local models via llama.cpp
  • Use RAG over your own material

For Developers building or using terminal-based LLM agents

Role
agent-app
Language
Python
Licence
MIT
Forks
431
Open issues
6
Last push
2026-09-17
Latest release
v0.2.1 · 2023-09-05
topicsllm-agentcliterminalcode-generationautonomous-agentsmulti-provider