#2020th of 28 in Agent platforms
openfang
Rust agent runtime with scheduled autonomous agents and 40 chat channel adapters
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 18.2k
- License
- Apache-2.0
- Last commit
- May 2026
- Last release
- May 2026
- Language
- Rust
Missed the latest GitHub check; these facts may be older.
Overview
OpenFang is a Rust agent runtime that ships as a single ~32MB binary with a web dashboard on port 4200. It bundles seven preconfigured autonomous agents ("Hands") such as Researcher, Lead and Browser, which run on schedules and report through 40 messaging adapters. It routes to 27 LLM providers, exposes an OpenAI-compatible API, and runs tool code in a WASM sandbox.
Who it is for: Self-hosters wanting scheduled autonomous agents that report via chat apps
Strengths
- Single binary with built-in dashboard, SQLite memory and vector embeddings
- 40 channel adapters, including Telegram, Slack, Discord, WhatsApp, Signal and Matrix
- 27 LLM providers including Ollama, vLLM and LM Studio, with fallback and cost tracking
- WASM tool sandbox, approval gates and a Merkle hash-chain audit trail
Weaknesses
- Pre-1.0; README warns of breaking changes and advises pinning to a commit
- README says MIT while repository license is Apache-2.0; version numbers also conflict
- Benchmark comparisons are self-reported from public docs, not independently verified
- WhatsApp QR gateway needs Node.js 18+ and a separate process
What it needs
- no GPU
- Docker + Compose
- Needs Node.js >= 18 (WhatsApp Web gateway only), FFmpeg and yt-dlp (Clip Hand), Playwright (Browser Hand)
- Models: Anthropic, Gemini, OpenAI-compatible, Ollama, vLLM
- port 4200
Also in Agent platforms
See all 28| Rank | Project | Score |
|---|---|---|
| 1 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 81 out of 100 |
| 2 | daily_stock_analysisDaily LLM stock analysis for A-share, HK, US and other markets, pushed to chat apps | 80 out of 100 |
| 3 | mindshubAgent workspace for knowledge work, with swappable open agent harnesses | 78 out of 100 |
| #4 | deer-flowAgent harness that runs sub-agents, memory and sandboxed skills for long tasks | 76 out of 100 |
| #5 | DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs | 75 out of 100 |