“Imitation is the sincerest form of flattery that mediocrity can pay to greatness.”— Oscar Wilde
Finds GitHub projects that look like AI-built (“vibe coded”) reimplementations of existing software, and shows the evidence for each one.
Loading the latest scan…
How to read this.AI-built counts agent config files (CLAUDE.md, AGENTS.md, .cursorrules…),
AI co-author trailers in recent commits, burst or big-bang commit history and typical README style.
Reimplementation counts claims like “drop-in replacement”, “clean-room”, “rewrite of” and “compatible with”.
The combined score is the geometric mean of the two, so a repo needs both to score high. Original chips appear when you
name an original repo. A changed license or a high file-name overlap is a reason to check provenance by hand, not proof of copying.
The scores are heuristics: a repo with no signals isn't necessarily human-written, and one with many signals isn't necessarily a legal problem.