explore-code Skill for Claude: Candidate Code Changes on a Branch

explore-code makes your agent try candidate code changes on an isolated branch of an ML repo, with rollback notes and no verified-result claims.

Installs: 344.7K; on skills.sh
Stars: 486; on GitHub
Files: 4; besides SKILL.md
Updated: Sep 2026; last repo commit

explore-code is the implementation leaf of the RigorPilot explore lane. Use it when you explicitly authorize exploratory work on an isolated branch or worktree, such as module transplant, backbone adaptation, LoRA or adapter insertion or replacing a head. It favors copying from a source and minimal adaptation over freeform rewrites.

It writes explore_outputs/ files: CHANGESET.md, SCIENTIFIC_CHANGELOG.md, COMPARABILITY_REPORT.md, TOP_RUNS.md and status.json, recording why a change matters, how to roll it back and why it is only a candidate. It can hand runs to minimal-run-and-audit or run-train, and it reads shared files from ai-research-reproduction, so install the whole set.

Good for

Researchers who want to transplant a module or swap a backbone or head
Engineers adding LoRA or adapter layers to an existing model
Anyone who needs each candidate change logged with a way back

Install explore-code

Claude Code

Copy this prompt.

Install the explore-code 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.

This skill reads shared guideline files from the ai-research-reproduction skill, so install the whole set rather than this folder alone. The README says the main reproduction skill works alone and that "all skills" is the choice for the companion and leaf skills.

From the project README. Source.

What's in the skill

When it loads

Rigor Improve implementation leaf skill for auditable candidate implementation in deep learning research repositories. Use when the researcher explicitly authorizes exploratory work on an isolated branch or worktree to transplant modules, adapt a backbone, add LoRA or adapter layers, replace a head, or stitch together meaningful low-risk migration ideas with rollback-aware records in `explore_outputs/`. Do not use for end-to-end exploration orchestration on top of `current_research`, trusted baseline reproduction, conservative debugging, environment setup, verified contribution claims, or default repository analysis.

Sections in SKILL.md

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

Key facts

Author: lllllllama (owner of the lllllllama/rigorpilot-skills repo)
License: MIT
Installs on skills.sh: 344,686 as of 2026-11-02
GitHub stars: 486 for the whole lllllllama/rigorpilot-skills repo
skills.sh audits: Gen Agent Trust Hub warn, Socket pass, Snyk pass
What is inside: 4 files besides SKILL.md (2 in scripts/, 1 in agents/, 1 in references/)
Needs: The other RigorPilot skills for shared reference files; Python for the bundled helpers
Last commit to the repo: 2026-09-23

What it includes

Keeps changes off the trusted baseline
Works only on an isolated branch or worktree that you authorized.
Try: “Use explore-code on a new branch: swap the classification head for an adapter layer.”
Runs code · Isolate the work
Copies from a source and adapts minimally
Covers module transplant, backbone adaptation, LoRA or adapter insertion and head replacement. It is not for from-scratch idea implementation or broad refactors.
Write · Source-anchored changes
Plans the code changes
Bundled Python script named in SKILL.md notes. SKILL.md does not say what it does.
Runs code · scripts/plan_code_changes.py
Why it matters, how to undo it
Writes CHANGESET.md and the scientific changelog. SKILL.md says they record why a change matters and how to roll it back.
Write · Rollback-aware record
Runs go to minimal-run-and-audit or run-train
explore-code itself owns code changes only. Use ai-research-explore when you need the whole campaign coordinated.
Runs code · Hand off execution

From the official docs

Permissions

Medium risk

The agent edits code in your repo on an isolated branch. Confirm the branch before it starts.

What it can do

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

Watch out for

It edits your code on a branch
SKILL.md requires explicit authorization and isolation from the trusted baseline. The README says approved commands can access the host and network and to use trusted repositories. Check you are on the right branch first.
Results are candidates, not verified
The skill records every change as a candidate. Do not cite it as a verified contribution.

Technical details

Scopes	Edits code on an isolated branch or worktree Bundles two Python scripts (plan_code_changes.py, write_outputs.py) that SKILL.md names Writes explore_outputs/ files

Questions

What does explore-code do?

It makes exploratory code changes to an ML repo on an isolated branch, then records what changed, how to roll back and why it is only a candidate.

How do I install it?

Use the install box at the top of this page, which installs the whole RigorPilot set. The skill reads shared files from ai-research-reproduction, so installing it alone is not enough. The README says the installer needs Node.js/npm and that skills@1.5.26 requires Node 22.20.0 or newer.

Does it need an API key?

No key is named in SKILL.md.

When does it not apply?

Not without your explicit authorization, and not for trusted baseline work, conservative debugging, normal training or broad refactors. SKILL.md sends whole campaigns to ai-research-explore.

explore-code or explore-run?

explore-code owns exploratory code changes only. explore-run owns exploratory execution planning and summary. ai-research-explore coordinates both.

Is it safe?

It edits code, so keep it on a branch you can discard; SKILL.md requires explicit authorization and isolation from the trusted baseline. The README says RigorPilot is local execution, not an OS sandbox. The folder also bundles two Python scripts.

Information

updated Nov 2, 2026

Repository: lllllllama/rigorpilot-skills
Folder: skills/explore-code

lllllllama/rigorpilot-skills skills.sh

Sources

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

Show all 4

RigorPilot Skills README · github.com
explore-code SKILL.md · github.com
explore-run SKILL.md · github.com
skills.sh listing and audits · skills.sh