lthoangg

lthoangg / openagentd

Public

Self-hosted AI agent OS — streaming chat, tool use, persistent memory, and multi-agent teams. Runs entirely on your machine.

github.comlthoanggopenagentd agents ai llm multi-agent self-hosted
58
5
100% credibility
Found May 02, 2026 at 13 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

OpenAgentd is a self-hosted multi-agent AI system featuring a web-based cockpit, persistent editable memory, built-in tools for file and shell operations, and support for team coordination across various AI providers.

How It Works

1
💡 Discover your local AI team

You hear about OpenAgentd, a helpful AI assistant that runs entirely on your computer with a friendly web dashboard.

2
📦 Install in seconds

Run a simple command to get it set up on your Mac, Linux, or Windows machine—no hassle.

3
🔑 Connect your favorite AI

Pick from popular AI services like Google Gemini or OpenAI and link it with your account details.

4
🚀 Launch the web cockpit

Open your browser to a clean dashboard with chat, file uploads, and agent controls.

5
💬 Chat and assign tasks

Type a request, drag in files or images, and watch agents read files, search the web, or run commands.

6
👥 Form a dream team

Set up a lead agent with helpers for research, building, or planning—they message each other to get complex work done.

✅ See real results

Your tasks complete locally with full privacy, editable memory, and everything saved in your workspace.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 13 to 58 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is openagentd?

OpenAgentd is a self-hosted AI agent OS built in Python with FastAPI and React, delivering a web cockpit for streaming chat, tool use, persistent memory, and multi-agent teams—all running locally on your machine. Drop in files, run shell commands, search the web, or manage todos via agents that coordinate via async mailboxes, with editable wiki memory spanning sessions. Install via `uv tool install openagentd` or Docker Compose, init providers like OpenAI or Gemini, then hit `openagentd` for http://localhost:4082.

Why is it gaining traction?

It beats CLI-only alternatives with a full web UI (command palette, file workspace, OTel dashboard), editable three-tier memory (user notes to synthesized wiki), and hot-reload configs—no restarts needed. Multi-agent teams delegate via `team_message`, schedules run cron jobs, and 12 providers (Gemini, Grok, Copilot) with fallbacks mean no lock-in. Self-modifying skills and MCP tool servers extend it without code.

Who should use this?

Backend devs rigging self-hosted agent pools for Azure DevOps or GitHub Actions runners with AI automation. Solo coders wanting a local, self-hosted GitHub Copilot alternative for agentic coding tasks like debugging or research. Teams experimenting with self-hosted agentic AI for secure, offline workflows.

Verdict

Promising beta for self-hosted agent AI, but 1.0% credibility score (13 stars) signals early days—docs and Docker install shine, though test coverage needs growth. Grab it if cloud agents irk you; skip for production until stars climb.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.