RigorPilot Bundle · 5 skills

Use it when: you want an agent to run a deep learning research repo from its README and record honestly what worked
It does: finds the smallest documented run, such as one evaluation command, sets it up, runs it and audits the result, instead of a full training run
You get: a repro_outputs/ folder: a verdict (verified, partial or blocked), the exact commands, logs, patches, and any change to data, evaluation or metrics
Main skill: ai-research-reproduction runs the steps. It isn't one of the five listed, but the bundle's install command adds it, and two of the five need it
Also has: ai-research-explore, a separate lane for trying new research ideas, whose results are never reported as reproduced

TLDR

The repo's main skill, ai-research-reproduction, runs a reproduction end to end. It isn't listed among these five, but the bundle's install command adds it with them. The five are its steps, and each can also be called on its own. repo-intake-and-plan reads the README and project files, sorts the documented commands into inference, evaluation and training, and picks the smallest one worth trusting. paper-context-resolver steps in only when the README leaves out a detail that matters, such as which dataset split to evaluate on, and looks it up in the paper, noting any conflict. env-and-assets-bootstrap drafts conda commands and where checkpoints and datasets should go, before anything runs. minimal-run-and-audit runs the chosen command and writes the evidence to repro_outputs/, including any patch and any change to data, evaluation or metrics; its scripts need the main skill. ai-research-explore, which needs it too, is a separate lane: only when you ask, and starting from a model or run you already have, it ranks and tries new research ideas, writes them to explore_outputs/, and never reports them as reproduced results.

Summarizes the repo at fb3ccdf.

Any agent, with the skills CLI

npx skills add https://github.com/lllllllama/rigorpilot-skills --skill '*'

Source on GitHub · All skills from lllllllama/rigorpilot-skills

Installs: 450,930 of paper-context-resolver, its most installed skill, on skills.sh, Sep 21, 2026
Repo stars: 492 lllllllama/rigorpilot-skills
First seen: Mar 30, 2026 on skills.sh, its first skill

Its skills, in the order they're used

The ones marked “Only in the bundle” need the rest of it, so they're only listed here. The others also have their own entry in the skills list.

repo-intake-and-plan
450K installs

Maps a deep-learning repo and plans a README-first reproduction.
paper-context-resolver
451K installs

Fills the gaps a README leaves when reproducing a deep-learning paper's repo.
env-and-assets-bootstrap
450K installs

Prepares a conda environment, datasets and checkpoints for a reproduction.
minimal-run-and-audit
Only in the bundle 450K installs

Runs a minimal reproduction and records the evidence.

Its run and report scripts load a shared runtime that ships inside ai-research-reproduction, the repo's main skill, and stop with an error without it.
ai-research-explore
Only in the bundle 311K installs

Explores novel research candidates on top of a reproduced baseline.

Loads its required research loop and shared runtime from ai-research-reproduction, the repo's main skill, and hands code changes and trial runs to the repo's explore-code and explore-run skills.