minimal-run-and-audit
๐ฏSkillfrom lllllllama/ai-paper-reproduction-skill
Executes and audits the selected smoke test, documented inference, or evaluation command during README-first AI paper reproduction, writing standardized `repro_outputs/` evidence and patch notes.
Overview
A sub-skill in the ai-paper-reproduction-skill repository that executes and audits the selected smoke test, documented inference, or evaluation command during README-first AI paper reproduction. Writes standardized repro_outputs/ evidence and patch notes.
Key Features
- Runs the selected smoke test, inference, or evaluation command
- Normalizes execution evidence into
repro_outputs/ - Produces auditable patch notes for any minimal repo changes
- Preserves the "trusted by default" lane โ no unnecessary modifications
Who is this for?
Researchers and agents reproducing AI papers who want standardized, auditable evidence instead of ad-hoc logs. Useful for creating reproduction reports that reviewers or collaborators can trust without re-running everything themselves.
Same repository
lllllllama/ai-paper-reproduction-skill(12 items)
Installation
npx vibeindex add lllllllama/ai-paper-reproduction-skill --skill minimal-run-and-auditnpx skills add lllllllama/ai-paper-reproduction-skill --skill minimal-run-and-audit~/.claude/skills/minimal-run-and-audit/SKILL.mdSKILL.md
More from this repository10
Companion skill for AI paper reproduction that resolves paper context, references, and dependencies needed when reproducing AI research implementations.
Helper skill for README-first AI paper reproduction that scans a repository, extracts documented commands, and returns the smallest trustworthy inference, evaluation, and training plan.
Sub-skill for README-first AI paper reproduction that prepares a conservative conda-first environment plus checkpoint, dataset, and cache path assumptions before any run attempt.
README-first AI paper reproduction skill that helps reproduce AI papers by choosing the smallest trustworthy documented target with minimal, auditable code changes.
A lane-aware skill repository for deep learning research workflows that separates trusted reproduction tasks from exploratory candidate work, shipping 11 skills with 42 test scripts across Windows and Linux for use with Claude Code, Codex, and Agent Skills.
Part of the AI research workflow skills collection, this skill analyzes research project structure and codebase to understand components, dependencies, and architecture before reproduction.
An explore-lane skill for making isolated, candidate-only code changes during deep learning research, requiring explicit researcher authorization and producing disposable outputs that don't affect trusted reproduction artifacts.
Part of the AI research workflow skills collection, this skill enables safe exploratory trial runs in a trusted lane model for reproducing and experimenting with AI research papers.
Part of the AI research workflow skills collection, this skill safely diagnoses tracebacks and failed training or inference runs with safety guardrails to prevent data loss.
Part of the AI research workflow skills collection, this skill manages training runs with safety guardrails in a trusted lane model for reproducing AI research experiments.