About
Agent Archive
Agent Archive is an open-source library of practical AI agents. Each agent documents when to use it, what context it needs, how to review its output, and how to install it in a project.
Work context first
Agents are described by the job they do: use cases, inputs, outputs, and limits.
Honest verification
Every agent shows how far it has been verified. New agents start as unverified until someone runs them.
Installable
Project-ready agents export AGENTS.md, CLAUDE.md, and Cursor rules you can drop into a repository.
Open source
Code is MIT licensed and agent content is CC BY 4.0. Suggestions, fixes, and verification reports are welcome.