Free · browser-only · no account

Is your repository ready for an AI coding agent?

Choose a local project folder to check ten practical readiness signals: repository instructions, project policy, hooks, CI, tests, ownership, documentation, dependency locking, and secret hygiene.

Your repository is not uploaded.

This page reads a bounded set of configuration and documentation files locally in your browser. It sends only the aggregate score and file count for funnel measurement—never filenames or file contents.

Run the readiness snapshot

Select the repository folder Chrome, Edge, and Safari support folder selection. Nothing is changed. Bounded to 3,000 paths and 2 MiB of relevant text. Secret files and project source are not read.

What the free audit checks

The snapshot looks for evidence that an agent can understand the repository, operate within explicit boundaries, and verify its work. A missing signal is a review prompt—not proof that the repository is unsafe.

Version-controlled CLAUDE.md or AGENTS.md instructions
Concise, actionable instruction content
Shared Claude Code or Codex project policy
Configured hooks or protected-path controls
Continuous-integration workflows and test evidence
CODEOWNERS, README, dependency lockfile, and ignored secret files

Turn the gaps into a working setup

Start with the smallest useful fix.

The $9 setup kit supplies seven tested hooks, repository instructions, permission profiles, CI examples, and an adaptation playbook. If you need a deeper evidence trail, the $49 local analyzer adds HTML, Markdown, and JSON reports, four operating profiles, remediation drafts, and the complete kit.

Open a sample report Download Doctor free View the open-source audit