There’s a new open-source project making serious noise in the agent space, and honestly, it deserves the attention. [OpenFang](https://github.com/RightNow-AI/openfang) launched on February 24th and racked up over 7,000 GitHub stars in under a week. By March 1st, it was sitting at [#2 on Trendshift](https://trendshift.io/) and pulling [143 upvotes on Product Hunt](https://www.producthunt.com/) on the same day. That kind of simultaneous traction across platforms doesn’t happen often.
So what is it? OpenFang calls itself an “Agent Operating System,” and for once, the name isn’t hyperbole. Built entirely in Rust by [Jaber at RightNow AI](https://www.rightnowai.co/), it compiles 137,000 lines of code across 14 crates into a single ~32MB binary. You download one file, run one command, and your agents are live. Cold start is around 180ms, idle memory sits at 40MB. Those are real numbers, not marketing fluff.
The core concept revolves around what they call “Hands” — seven pre-built autonomous agents that handle things like lead generation, OSINT monitoring, video content processing, research with source credibility scoring, and even Twitter account management. These aren’t chatbot wrappers. They run on schedules, build knowledge graphs, and report back to a dashboard without you poking them every five minutes.
What really caught my eye is the integration breadth. Forty channel adapters covering everything from Telegram and Discord to WhatsApp, Signal, Matrix, and Microsoft Teams. Twenty-seven LLM providers supported natively, including Anthropic, Gemini, and any OpenAI-compatible endpoint. Plus 38 built-in tools and MCP support if you want to extend further.
The security story is unusually thorough for an open-source project at this stage. We’re talking WASM sandboxing with dual metering, Ed25519 manifest signing, Merkle hash-chain audit trails, prompt injection scanning, taint tracking, and SSRF protection — sixteen security layers in total. That’s the kind of setup you’d expect from a funded enterprise product, not a v0.1.0 release.
It’s MIT-licensed, so you can do whatever you want with it. The [official site](https://www.openfang.sh/) has a clean overview, and there’s a Tauri 2.0 desktop app if you prefer a GUI. The project also showed up on [bestofshowhn.com](https://bestofshowhn.com), which tracks the best Hacker News launches.
Is it production-ready? The team is targeting a solid v1.0 by mid-2026, so treat it accordingly. But as a foundation for building real autonomous agent workflows — not toy demos — OpenFang is the most complete package I’ve seen ship as a single binary. Worth keeping on your radar.

Leave a comment