Check whether your repositories are ready for AI agents

Coding agents are only as effective as the system they work in. They need two things from a repository: configuration files like AGENTS.md or CLAUDE.md that tell them how to build, test, and navigate the codebase, and a delivery pipeline that can take their output once they’re done. Engineers open more pull requests with agents, which means CI runs more often. Slow CI now taxes every pull request an agent opens, not just every engineer’s.

The AI readiness view shows whether each of your active repositories is configured for your agents and how long it takes for merged work to reach production, so you can find the busy repositories where agents lack context or where changes are slow to ship.

From there, you can open the pull requests that last updated a repository’s configuration files and check that the instructions are still current. If a repository’s time to deploy stands out, go to CI visibility to find the slow or failing workflows behind it.

To get started, go to AI tools → AI readiness.