Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLIAbout this skill
Evidence-driven ML repository verification skill. Treats every Claude Code claim as a hypothesis to test, not a statement to trust. Runs staged audits (quick / standard / forensic) across code executability, data reality, training authenticity, and metric validity. Produces claim verdicts, findings JSON, evidence manifest, and a human-readable audit report. Use before merging ML code, before model releases, or whenever AI-generated code claims need independent verification. Triggers on: "audit ml repo", "verify claims", "audit this repo", "check training is real", "verify metrics", "run quick audit", "run standard audit", "forensic audit", "hallucination check", "evidence-driven audit", "verify model training".
Quick start
- Copy the install prompt. Use the green button above this README.
- Send it to your agent. Paste it into your coding agent’s chat and send.
- Review the setup. Your agent checks the skill’s instructions, requirements, and installation scope.
Review, license & listing details
This is an imported listing. We haven’t certified its safety, license, or compatibility. Skills may include executable scripts.
Only SKILL.md was scanned. Supporting files and scripts were not checked. No safety grade has been assigned.
0 potential findings · Scanned 11/4/2026 · Scanner 0.2.1
Content SHA-256: c1e67201fc75ac005803803de04ea9db3c0edd002357e1f0ecb75670df1a4818
About this listing
- Category
- Science & research
- Reported license
- Not specified
- Listing language (estimated)
- English
- Review status
- Not individually reviewed
- Agent compatibility
- Atlas installation supports Claude Code, Codex, and Cursor. Skill-specific dependencies and features are checked during setup.