AI Repository Detector
A repository-wide view of AI-assisted code
Point AI Inspector Lab at a public GitHub repository, branch or sub-folder. We read repository metadata, sample source files across every directory, skip vendored and generated code, and present a repository-wide report with file and directory breakdowns.
What the inspector checks
Representative sampling
Files are sampled round-robin across directories so one large folder cannot dominate the result.
Smart exclusions
node_modules, vendor, build output, lockfiles, minified bundles, binaries and generated code are excluded and counted.
Directory breakdown
See which parts of a codebase concentrate assistant-style patterns or higher model scores.
No credentials needed
Public repositories are read through GitHub’s public API; nothing is cloned or executed.
Who uses it
- Open-source due diligence
- Evaluating vendor codebases
- Hackathon and competition judging
- Engineering leadership reviews
AI Repository Detector FAQ
Do I need to connect my GitHub account?
No. Public repositories are fetched anonymously (or with the server’s configured token for higher rate limits).
Is the code executed or installed?
Never. Files are downloaded as text and analysed statically.
How big can a repository be?
Any size can be submitted; up to 200 representative source files are analysed and the report states how many were sampled.
Other detectors
Try the AI Repository Detector for free
Check text, public websites and GitHub repositories for free - no account needed. Go Pro for private document, image and code inspections, Compare and the API.