# 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!

Repository: https://github.com/gptme/gptme
Canonical: https://ross.abutalabs.com/products/gptme
Homepage: https://gptme.org/docs/
Language: Python
License: MIT
License Family: permissive
Topics: chatbot, chatgpt, llamacpp, llm, openai, cli, ai-agents, ai-assistant, code-generation, llm-agent, llm-apps, rag, anthropic, openrouter, agent, agents
Last push: 2026-08-26T22:07:18+00:00

## Health v2 (maintenance only)
Score: 97/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 98, longevity 89
- inputs: {"age_days": 1258, "days_push": 7, "days_rel": 14, "gap_med": 8, "n_releases_24m": 28}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4399, forks 418 (observed 2026-08-28T04:08:48.248930+00:00)

## What it is
gptme is a free, open-source personal AI agent that runs in the terminal, equipped with local tools for shell execution, code writing, file editing, web browsing, and vision. It is provider-agnostic (Anthropic, OpenAI, local llama.cpp, etc.) and can also be used as a library, server, or foundation for building custom persistent agents.

## Use cases
- run an AI coding agent in my terminal
- have an LLM fix failing tests and commit changes
- automate code review on pull requests in CI
- chat with an AI that can execute shell commands and edit files
- build my own autonomous agent on top of an existing framework
- use a local LLM agent without sending data to the cloud
- let an AI browse the web and summarize pages from the terminal

## When to choose
- you want a terminal-first, provider-agnostic coding agent as an alternative to Claude Code or Codex
- you need local-first operation with local models via llama.cpp
- you want to automate AI workflows in CI/CD pipelines or scripts
- you want to build custom persistent agents on an extensible platform with MCP support

## When to avoid
- you need a polished GUI-first IDE experience like Cursor
- you primarily work on Windows natively without WSL or Docker
- you want a fully hands-off agent without any confirmation or oversight

## Facets
- artifact type: cli-tool
- maturity: active
- function: agent-framework, chatbot, cli, rag, llm-inference, web-scraping, code-review
- domain: large-language-models, developer-tools, chatbots
- platform: cli, python, cross-platform
- tags: terminal-agent, coding-agent, mcp, local-first, provider-agnostic, computer-use, subagents, ai-agents, command-line, automation, linux, macos

## Member repositories
- gptme/gptme (main) score 97

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:48.248930+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:21:09.190871+00:00, confidence not recorded.
  - readme: https://github.com/gptme/gptme (fetched 2026-08-28T04:08:48.248930+00:00, sha dfbac5d242dc)
  - homepage: https://gptme.org/docs/ (fetched 2026-08-29T09:09:13.274219+00:00, sha d73aa70be43b)
  - site_page: https://gptme.org/docs/getting-started.html (fetched 2026-08-29T09:09:13.277330+00:00, sha b7763a64a8ed)
  - site_page: https://gptme.org/docs/system-dependencies.html (fetched 2026-08-29T09:09:13.279252+00:00, sha ff53768e9a07)
  - site_page: https://gptme.org/docs/usage.html (fetched 2026-08-29T09:09:13.281003+00:00, sha fad8084cb3be)
  - site_page: https://gptme.org/docs/examples.html (fetched 2026-08-29T09:09:13.282817+00:00, sha 308cdb44df98)
  - site_page: https://gptme.org/docs/demos.html (fetched 2026-08-29T09:09:13.284755+00:00, sha 0e6084cf8cb7)
  - site_page: https://gptme.org/docs/automation.html (fetched 2026-08-29T09:09:13.286619+00:00, sha 89050b797233)
  - site_page: https://gptme.org/docs/projects.html (fetched 2026-08-29T09:09:13.289470+00:00, sha 9fd046ed2696)
  - site_page: https://gptme.org/docs/cookbook.html (fetched 2026-08-29T09:09:13.291236+00:00, sha 4b3060877c96)
- Data as of 2026-08-30T08:39:29.467469+00:00.