AI BriefingTrueForge, Open-Source Agent Harness Released
Key point
TrueFoundry has released TrueForge, an open-source harness that integrates the agent execution loop from model calls to sandboxes and UI.
Details
TrueFoundry has released TrueForge, an open-source harness for AI agent prototyping and operations. This tool provides integrated infrastructure required for agent execution, including model calls, MCP tool integration, code execution sandboxes, user approval, and session management.
TrueForge is served in three ways: Chat UI, HTTP API (TypeScript SDK), and UI SDK, making it suitable for various use cases such as internal team use, automation pipelines, and product embedding. It supports models from OpenAI, Anthropic, Google Gemini, and OpenAI-compatible endpoints, and uses Daytona for sandboxes.
According to its own benchmark results, it achieved the same accuracy as Claude Managed Agents while being approximately 30% cheaper, and 2.5x cheaper than the existing open-source harness deepagents. This is due to reduced token usage through context compression and offloading of large results.
- License: MIT
- Requirements: Node.js 22.14 or higher
- Deployment: Local mode (SQLite) or Hosted mode (Postgres + Redis)
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Whiteboard Launches Open-Source VS Code Fork for Human-AI Collaborative Software Design
OpenComputer Hosts 'Ship Video' Agent Using Claude Opus 5.5 to Generate Explainer Videos from Code