Env and Assets Bootstrap Skill: Set Up a Repro Env

Plans a conda-first environment plus checkpoint and dataset paths before you run a research repo.

Installs: 484K; on skills.sh
Stars: 486; on GitHub
Files: 7; besides SKILL.md
Updated: Sep 2026; last repo commit

env-and-assets-bootstrap is the setup step of README-first deep learning reproduction. It prepares a conservative conda-first environment plan, candidate conda commands, an asset path plan, checkpoint and dataset source hints, and a list of unresolved dependency or asset risks.

Use it after repo intake has found a credible target and before any run. It is not for repo scanning, final reporting, paper lookup or generic conda questions. It bundles four scripts (bootstrap_env.py, bootstrap_env.sh, plan_setup.py, prepare_assets.py) and two policy notes, and it reads shared principles from the sibling ai-research-reproduction skill.

Good for

Researchers getting a paper repo to run on their machine
ML engineers who need checkpoint and dataset paths planned first
Users of the RigorPilot main skill

Install env-and-assets-bootstrap

Claude Code

Copy this prompt.

Install the env-and-assets-bootstrap skill in Claude Code. Run this command:

npx skills add lllllllama/RigorPilot-Skills --all

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js 22.20+.

The README says the main skill, ai-research-reproduction, works alone, and to choose all skills for the companion entrypoints. SKILL.md reads shared files from the ai-research-reproduction skill, so install the set. The README tested the installer with skills@1.5.26 and Node 22.20.0, which it says is the minimum.

From the project README. Source.

What's in the skill

When it loads

Rigor Setup skill for README-first deep learning repo reproduction. Use when the task is specifically to prepare a conservative conda-first environment, checkpoint and dataset path assumptions, cache location hints, and setup notes before any run on a README-documented repository. Do not use for repo scanning, full orchestration, paper interpretation, final run reporting, or generic environment setup that is not tied to a specific reproduction target.

Sections in SKILL.md

When to apply
When not to apply
Clear boundaries
Input expectations
Output expectations
Notes

Key facts

Author: lllllllama (RigorPilot-Skills)
Installs on skills.sh: 484,009 (Nov 2, 2026)
GitHub stars: 486 on lllllllama/rigorpilot-skills (Nov 2, 2026)
License: MIT
Last repo commit: Sep 23, 2026
What installs: SKILL.md plus 7 bundled files, mostly reference notes under references/
skills.sh audits: Gen Agent Trust Hub: pass, Socket: pass, Snyk: pass
Part of: RigorPilot-Skills; reads shared principles from the ai-research-reproduction skill
Runs: scripts/plan_setup.py, bootstrap_env.py, bootstrap_env.sh and prepare_assets.py
Reference: references/env-policy.md and references/assets-policy.md

What it includes

Plans environment and asset assumptions
Uses scripts/plan_setup.py and the README setup steps.
Runs code · Plan setup
Prepares a conservative conda-first environment
Uses scripts/bootstrap_env.py, or bootstrap_env.sh as a POSIX wrapper.
Runs code · Bootstrap the environment
Plans checkpoint, dataset and cache paths
Uses scripts/prepare_assets.py. Gives source hints for checkpoints and datasets.
Runs code · Prepare assets
Lists unresolved dependency or asset risks
Returned with candidate conda commands and setup notes.
Report risks
Passes paper questions to the optional paper resolver
Does no paper lookup itself.
Forward paper gaps

From the official docs

Permissions

Medium risk

It runs bundled setup scripts that can create environments and asset folders. Review the setup plan first and ask it to confirm before large downloads.

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

It runs setup scripts that change your machine
Environment bootstrap and asset preparation can create conda environments and folders. Read the plan before you let it apply.
Asset setup can involve large files
SKILL.md plans checkpoint, dataset and cache paths. The RigorPilot README example prompt tells the agent to ask before large downloads or long training, so keep that line in your prompt.

Technical details

Talks to	Not stated in SKILL.md. The bundled setup scripts may fetch packages or assets, so read them first.
Scopes	Creates environment and asset directories Runs bundled setup scripts

Questions

What does env-and-assets-bootstrap do?

It prepares a conservative conda-first environment plan and asset paths for a README-documented repo: candidate conda commands, checkpoint and dataset path assumptions, cache hints and risk notes.

How do I install it?

Follow the install box at the top of the page. It reads files from the sibling ai-research-reproduction skill, so install the whole RigorPilot set.

When do I use it?

After repo intake finds a credible reproduction target and before you run any command.

Does it run the experiment?

No. Running and reporting belong to minimal-run-and-audit. This skill stops at setup.

Is it for any conda setup?

No. SKILL.md says not to use it for generic environment setup that is not tied to a specific reproduction target.

Is it safe?

It runs setup scripts that can create environments and folders. Read the plan first, use trusted repos, and ask it to confirm before big downloads. Its skills.sh audits show pass from all three auditors.

Information

updated Nov 2, 2026

Repository: lllllllama/rigorpilot-skills
Folder: skills/env-and-assets-bootstrap

lllllllama/rigorpilot-skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 3

RigorPilot-Skills README · github.com
env-and-assets-bootstrap SKILL.md · github.com
skills.sh listing · skills.sh