🎯

minimal-run-and-audit

🎯Skill

from lllllllama/ai-paper-reproduction-skill

What it does

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 — 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)

minimal-run-and-audit

Installation

Vibe Index InstallInstalls to .claude/skills/

npx vibeindex add lllllllama/ai-paper-reproduction-skill --skill minimal-run-and-audit

skills.sh Install⚠ Installs to .agents/skills/

npx skills add lllllllama/ai-paper-reproduction-skill --skill minimal-run-and-audit

Manual InstallCopy SKILL.md content and save to the path below

~/.claude/skills/minimal-run-and-audit/SKILL.md

SKILL.md

139,673Installs

AddedApr 13, 2026

View on GitHub Back to Skills

More from this repository10

🎯

paper-context-resolver🎯Skill

Companion skill for AI paper reproduction that resolves paper context, references, and dependencies needed when reproducing AI research implementations.

🎯

repo-intake-and-plan🎯Skill

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.

🎯

env-and-assets-bootstrap🎯Skill

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.

🎯

ai-paper-reproduction🎯Skill

README-first AI paper reproduction skill that helps reproduce AI papers by choosing the smallest trustworthy documented target with minimal, auditable code changes.

🎯

ai-research-reproduction🎯Skill

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.

🎯

analyze-project🎯Skill

Part of the AI research workflow skills collection, this skill analyzes research project structure and codebase to understand components, dependencies, and architecture before reproduction.

🎯

explore-code🎯Skill

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.

🎯

explore-run🎯Skill

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.

🎯

safe-debug🎯Skill

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.

🎯

run-train🎯Skill

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.