trueforge
The open-source agent harness - the runtime layer that turns an LLM into a working agent.
TrueForge is an open-source agent harness that provides the runtime layer to transform language models into fully operational AI agents. Developed by TrueFoundry, it handles the complete agent execution loop including model calls, MCP tool integration, skill management, code sandboxing, human approval workflows, context management, and session persistence. The platform exposes three interfaces: a built-in chat UI, an HTTP API with TypeScript SDK, and an embeddable UI component. It scales from local SQLite-based development to production Postgres deployments via Docker Compose or Kubernetes, supporting any OpenAI-compatible model provider and remote MCP servers with authentication.
Key Features
Use Cases
- 01Building production-ready AI agents with built-in UI and API without writing execution loop code
- 02Running multi-tool agents that require human approval checkpoints before executing sensitive operations
- 03Deploying team-shared agent instances with persistent sessions and OIDC authentication
- 04Integrating multiple MCP servers with OAuth flows for agents that need authenticated external tool access
- 05Creating agents with isolated code execution capabilities using the sandbox-as-a-tool pattern
- 06Embedding conversational AI interfaces into existing applications using the UI SDK component
Related Agents
View morehermes-agent
The agent that grows with you
agency-agents
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
openinterpreter
A coding agent for open models like Kimi K3
cline
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
trueforge — FAQ
What is TrueForge?+
TrueForge is an open-source agent harness that provides the complete runtime infrastructure to turn an LLM into a working agent. It manages the execution loop, MCP tools, code sandboxing, approvals, context, and sessions, exposing everything through a chat UI, HTTP API, and embeddable SDK.
How do I install TrueForge?+
For local development, run 'npx @truefoundry/trueforge@latest' which starts a SQLite-backed instance on localhost. For production teams, deploy using Docker Compose or Kubernetes Helm charts with Postgres and Redis backing stores.
Which AI model providers does TrueForge support?+
TrueForge supports OpenAI, Anthropic, Google Gemini, and other catalog providers, plus any OpenAI-compatible API endpoint. Models are configured once in catalogs and agents select from the available options.
Do I need API keys to use TrueForge?+
Yes, you need API keys for whichever model provider you choose (OpenAI, Anthropic, etc.). You may also need credentials for any MCP servers or sandbox providers you configure in your catalogs.
Is TrueForge free to use?+
Yes, TrueForge is released under the MIT License and is completely free and open-source. You only pay for the underlying services you use (LLM API calls, cloud infrastructure if hosted).
What is the difference between local and hosted mode?+
Local mode uses SQLite with no extra infrastructure, runs in a single process, and is intended for personal development on localhost only. Hosted mode uses Postgres and Redis, supports multiple replicas, and is designed for team deployments with optional OIDC authentication.
How do I install trueforge?+
Open the source repository on GitHub and follow its README. trueforge is a agent — MCP Agents Market links you directly to the official repo.
Is trueforge free?+
trueforge is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.