Analyze Project

When you clone a deep learning repo and need to understand what you're looking at before you break something, this is your read-only audit pass. It maps model architectures, training and inference entry points, config files, and flags suspicious patterns without touching code or spinning up expensive jobs. The scope discipline is the real value here: it won't try to fix bugs, refactor speculatively, or run your failing command, so you get orientation without scope creep. Outputs land in analysis_outputs as markdown summaries and a status JSON. Think of it as the mandatory reconnaissance step before you start adapting someone else's research code.

skills.sh installs: 353.6K
GitHub stars: 486
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add lllllllama/rigorpilot-skills --skill analyze-project --agent claude-code

Installs into .claude/skills of the current project.

Files

analyze-project

Use this as the Rigor Analyze / Rigor Audit read-only skill. The installed slug remains analyze-project for compatibility.

Use the shared operating principles in ../ai-research-reproduction/references/agent-operating-principles.md; this skill should guide read-only analysis without constraining the model's project-specific reasoning.

When to apply

The user wants to understand a deep learning repository before changing it.
The user needs a map of model structure, training entrypoints, inference entrypoints, and config relationships.
The user wants conservative suggestions about likely insertion points or suspicious implementation patterns.
The user explicitly wants read-only analysis and not heavy execution.