Agent Harnesses is an open-source curated and ranked catalog of more than 100 AI agent harnesses. It provides machine-readable JSON and llms.txt data plus an MCP server so developers and AI agents can discover and recommend tools.
In the Frameworks & runtimes space, Agent Harnesses takes a focused approach. Finding, comparing, and recommending suitable AI agent harnesses from a ranked, machine-readable catalog. It is built as an open-source project for AI developers and agents building or evaluating agent harnesses. The project is open source (Open Source). Agent Harnesses is available on the web and the command line, and it can be self-hosted.
Ryan Alberts builds and maintains Agent Harnesses. Key capabilities include ranked catalog, weekly rescoring, and MCP server. It exposes integrations via an MCP server.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do