TrueFoundry, a San Francisco B2B machine learning startup, has released its custom TrueForge harness under the permissive MIT License on Github. This move aims to solve a growing enterprise problem as AI agents proliferate: enabling greater developer control of agents and tools while reducing cost.
The company states that when it used TrueForge paired with the open source GLM-5.2 LLM to complete 11 of 14 tasks on DevRev’s Enterprise-Bench, it cost 75% less than achieving the same results with Anthropic’s Claude Managed Agents harness powered by Claude Opus 4.8.
Using the same model in each harness, Opus 4.8, TrueFoundry still claims a cost savings of roughly 30% using TrueForge compared to Claude Managed Agents. The cost was $8.50 vs $11.80.
Anuraag Gutgutia, TrueFoundry’s co-founder and COO, said the company is giving away the harness for free because customers have been asking for it. “You have an ability where you bring in agents and MCPs — can we also get something where you can actually launch these managed agents?” he said.
TrueForge’s architecture centers on context engineering — controlling how much information gets sent back into the model on every step of an agent run. This includes delaying the loading of MCP tool schemas until they are needed and delegating isolated tasks to subagents.
The documentation sets the default compaction threshold at 50,000 tokens, though it can be changed per agent. TrueForge also treats the sandbox differently from runtimes that keep an agent inside an isolated environment throughout its run.
For enterprise developers, the practical proposition is that they can start locally with a single command and SQLite, then move the same agent harness into a shared deployment using Docker Compose or Helm with Postgres and Redis. TrueFoundry explicitly warns that the local configuration is intended only for use on a developer’s machine, not as an internet-facing production service.
TrueForge compares to other leading orchestration harnesses like Claude Managed Agents. While Claude Managed Agents is a fully managed production agent runtime operated by Anthropic, TrueForge is an open-source, vendor-neutral harness.
The key differentiator for TrueForge is its combination of an open-source, vendor-neutral harness with a clear path from local development to a shared production runtime, plus an optional enterprise governance plane through TrueFoundry.
As AI modules continue to evolve, TrueFoundry’s move to release TrueForge as an open-source harness may have significant implications for the industry. By providing a cost-effective and flexible solution for enterprise developers, TrueFoundry is poised to play a major role in shaping the future of AI adoption.
One of the most important distinctions for enterprise buyers is between TrueForge by itself and TrueForge connected to TrueFoundry’s commercial AI Gateway. The open-source harness can run independently, but it does not magically inherit an organization’s enterprise access policies on its own.
Gutgutia said the company ultimately wants its AI Gateway to become the common layer beneath whichever agents and harnesses an enterprise chooses. This would allow for a centralized control plane that can enforce policies, monitor spend, and keep services available.
It has also been reported that the use of AI defense systems can greatly improve the security of enterprise networks.
