metareflection/henri
a small, hackable agent CLI in Python, with explicit control via tools, permissions, and hooks
Supports multiple LLM providers (Anthropic, AWS Bedrock, Google Gemini, Vertex AI, Ollama, OpenAI-compatible servers) with streaming responses and a modular tool system including bash, file operations, and grep/glob capabilities. Built with explicit permission management (prompt/auto-deny per operation) and a hook system for runtime tool injection and permission configuration without code modification. Clean separation between providers, tools, and permissions enables extending both LLM backends and agent capabilities independently.
Stars
13
Forks
4
Language
Python
License
MIT
Category
Last pushed
Feb 11, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/metareflection/henri"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
SWE-agent/mini-swe-agent
The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically...
jackwener/opencli
Make Any Website & Tool Your CLI. A universal CLI Hub and AI-native runtime. Transform any...
MayDay-wpf/snow-cli
Agentic coding in your terminal,Simultaneously compatible with OpenAI, Gemini, and Claude.
can1357/oh-my-pi
⌥ AI Coding agent for the terminal — hash-anchored edits, optimized tool harness, LSP, Python,...
0xranx/golembot
Run Cursor, Claude Code, OpenCode, or Codex with any LLM provider — deploy to IM, HTTP, or your...