Reason we opened TrueForge is most open harnesses out there do one of these 2 things:
1. Either a CLI coding agent that's great for individual devs
2. A framework that you assemble during runtime
That's the gap we're looking to solve. TrueForge provides full runtime. Build agents via UI/APIs, run them self-hosted with an out-of-the-box sandbox, MCP tools, approvals, session traces.
Also, we benchmarked ourselves against popular harnesses like Claude Managed Agents and deepagents; and we're significantly cheaper at similar accuracy because of the context engineering.
Would love it if you're check the repo. We're actively working on improving the core harness further in the coming weeks!
Hey, I'm from TrueFoundry team that open-sourced this. Pi is great but it's solving a different problem than we are.
It's a CLI coding agent that lives in your terminal, built mainly for devs working on a codebase on their own machines.
TrueForge is a runtime for building and running general agents. It comes with a server and web UI, plus an SDK and API. So you can build production agents and run them for yourself or your org, self-hosted behind SSO.
The idea is providing an end-to-end path: build agents in the UI, access them via API, and deploy for your team.
Can I bring my models from openrouter to start experimenting?
You can use OpenRouter via their OpenAI-compatible endpoint. TrueForge allows adding any OpenAI-compatible endpoint.
We support most of the popular frontier and open models directly + some models via inference platforms like Together and Fireworks.
Native OpenRouter support and more providers are coming soon.
Why use this instead of the other more well-established open-source agent harnesses out there?
Fair question.
Reason we opened TrueForge is most open harnesses out there do one of these 2 things: 1. Either a CLI coding agent that's great for individual devs 2. A framework that you assemble during runtime
That's the gap we're looking to solve. TrueForge provides full runtime. Build agents via UI/APIs, run them self-hosted with an out-of-the-box sandbox, MCP tools, approvals, session traces.
Also, we benchmarked ourselves against popular harnesses like Claude Managed Agents and deepagents; and we're significantly cheaper at similar accuracy because of the context engineering.
Would love it if you're check the repo. We're actively working on improving the core harness further in the coming weeks!
Especially Pi, that's kind of the gold standard for open harnesses
Hey, I'm from TrueFoundry team that open-sourced this. Pi is great but it's solving a different problem than we are.
It's a CLI coding agent that lives in your terminal, built mainly for devs working on a codebase on their own machines.
TrueForge is a runtime for building and running general agents. It comes with a server and web UI, plus an SDK and API. So you can build production agents and run them for yourself or your org, self-hosted behind SSO.
The idea is providing an end-to-end path: build agents in the UI, access them via API, and deploy for your team.
[dead]