PluginBench
Skill
Pass
Audit score 90

repo-intake-and-plan

lllllllama/ai-paper-reproduction-skill

README-first repository scanner for deep learning reproduction planning.

What is repo-intake-and-plan?

Scans a repository's README and project files to extract documented commands and classify reproduction candidates (inference, evaluation, training) without executing anything. Use at the start of a reproduction task to build a trustworthy plan before setup or execution.

  • Scans repository structure and reads README and common project files
  • Extracts documented commands and classifies them by type (inference, evaluation, training, other)
  • Generates a minimum trustworthy reproduction recommendation
  • Identifies ambiguities and risks in the reproduction plan
  • Returns structured output for orchestrator-level decision-making

How to install repo-intake-and-plan

npx skills add https://github.com/lllllllama/ai-paper-reproduction-skill --skill repo-intake-and-plan
Claude Code
Cursor
Windsurf
Cline

How to use repo-intake-and-plan

  1. 1.Provide the target repository path to the skill
  2. 2.Optionally specify priority hints (e.g., inference-first)
  3. 3.The skill scans the README and common project files
  4. 4.Review the returned repo structure summary and command inventory
  5. 5.Examine the classified candidates and minimum reproduction recommendation
  6. 6.Check the ambiguity and risk list before proceeding to setup or execution

Use cases

Good for
  • Starting a deep learning paper reproduction to map available commands before execution
  • Assessing a new repository to understand its structure and documented workflows
  • Planning which reproduction step (inference, evaluation, or training) to attempt first
  • Identifying gaps or unclear instructions in a repository's documentation
  • Feeding repo analysis to a higher-level orchestrator for next-step decisions
Who it's for
  • Researchers reproducing deep learning papers
  • AI agents orchestrating multi-step reproduction workflows
  • Teams evaluating repository quality and documentation completeness
  • Users wanting to inspect a repo before committing to setup or execution

repo-intake-and-plan FAQ

Does this skill execute commands or set up environments?

No. It only scans, reads, and plans. It does not install dependencies, download assets, or run reproduction commands.

What files does it read?

Primarily the README and common project files. It uses repo-scan-rules.md and helper scripts to guide the scan.

When should I use this instead of running commands directly?

Use this at the start of a reproduction task to understand the repo structure and available commands before committing to execution or setup.

What does 'minimum trustworthy reproduction' mean?

The smallest, most conservative set of documented steps that can be executed with confidence based on the repository's own documentation.

Can it handle repos without a README?

It is designed for README-first workflows. Repos without clear documentation will be flagged in the ambiguity and risk list.

Full instructions (SKILL.md)

Source of truth, from lllllllama/ai-paper-reproduction-skill.


name: repo-intake-and-plan description: Rigor Intake helper for README-first deep learning repo reproduction. Use when the task is specifically to scan a repository, read the README and common project files, extract documented commands, classify inference, evaluation, and training candidates, and return the smallest trustworthy reproduction plan to the main orchestrator. Do not use for environment setup, asset download, command execution, final reporting, paper lookup, or end-to-end orchestration.

repo-intake-and-plan

Use this as the Rigor Intake helper. The installed slug remains repo-intake-and-plan for compatibility.

When to apply

  • At the beginning of README-first reproduction work.
  • When the main skill needs a fast map of repo structure and documented commands.
  • When inference, evaluation, and training candidates must be classified conservatively.
  • When the user explicitly wants to inspect the repo first and not run anything yet.

When not to apply

  • When execution has already started and the task is now about running commands or writing outputs.
  • When the target is not a repository-backed reproduction task.
  • When the user only wants paper interpretation without repo inspection.
  • When the user already has a selected documented command and only needs setup or execution.

Clear boundaries

  • This skill scans and plans.
  • This skill is helper-tier and should usually be orchestrator-invoked.
  • It does not install environments.
  • It does not prepare large assets.
  • It does not execute substantive reproduction commands.
  • It does not decide high-risk patching.

Input expectations

  • Target repository path.
  • Access to README and common project files if present.
  • Optional user hints about desired priority, such as inference-first.

Output expectations

  • concise repo structure summary
  • documented command inventory
  • inferred candidate categories: inference, evaluation, training, other
  • minimum trustworthy reproduction recommendation
  • notable ambiguity or risk list

Notes

Use references/repo-scan-rules.md and helper scripts under scripts/.