Skill

explore-code

From lllllllama-ai-paper-reproduction-skill

Performs exploratory code adaptations in deep learning research repos on isolated branches: transplants modules, adapts backbones, adds LoRA/adapters, replaces heads. Records summaries in explore_outputs/.

Python

Git

ai-ml

git-workflow

npx claudepluginhub lllllllama/ai-research-workflow-skills

Tool Access

This skill uses the workspace's default tool permissions.

Preview

- When the researcher explicitly authorizes exploratory code changes on an isolated branch or worktree.

Supporting Assets

agents/openai.yamlreferences/explore-policy.mdscripts/plan_code_changes.pyscripts/write_outputs.py

SKILL.md

Similar Skills

ai-research-explore

Orchestrates end-to-end candidate-only AI research exploration atop current_research, with auditable repo understanding, idea gating, bounded code adaptation, and governed experiments to explore_outputs/.

20 files

lllllllama-ai-paper-reproduction-skill

🔬 Awesome Autoresearch

Curates autoresearch patterns for autonomous loops: LLMs propose code/ML changes, measure metrics, keep improvements or revert. Includes Python code and Claude skill setup.

aradotso-trending-skills-37

researcher

210

Orchestrates autonomous experiments to optimize measurable metrics like build time, latency, accuracy, or configs via git branches and .lab/ logging.

researcher

Stats

Stars10

Forks0

Last CommitApr 4, 2026

Actions

View Source View Plugin View on GitHub View README

Help us improve

Share bugs, ideas, or general feedback.

explore-code

When to apply

When the researcher explicitly authorizes exploratory code changes on an isolated branch or worktree.

When the task is source-anchored module transplant, backbone adaptation, LoRA or adapter insertion, or low-risk module combination.

When summary-level recording is sufficient and the result is a candidate, not a trusted conclusion.

When not to apply

When the request is for trusted baseline work, conservative debugging, or normal training execution.

When the user did not explicitly authorize exploratory modifications.

When the task is a broad refactor or a from-scratch idea implementation.

Clear boundaries

This skill owns exploratory code modifications only.

It must keep work isolated from the trusted baseline.

Use ai-research-explore instead when the task spans both current_research coordination and exploratory runs.

It may hand off execution to minimal-run-and-audit or run-train.

It should favor source-anchored copying and minimal adaptation over freeform rewrites.

Output expectations

explore_outputs/CHANGESET.md

explore_outputs/TOP_RUNS.md

explore_outputs/status.json

Notes

Use references/explore-policy.md, scripts/plan_code_changes.py, and scripts/write_outputs.py.

explore-code

When to apply

When the researcher explicitly authorizes exploratory code changes on an isolated branch or worktree.
When the task is source-anchored module transplant, backbone adaptation, LoRA or adapter insertion, or low-risk module combination.
When summary-level recording is sufficient and the result is a candidate, not a trusted conclusion.

When not to apply

When the request is for trusted baseline work, conservative debugging, or normal training execution.
When the user did not explicitly authorize exploratory modifications.
When the task is a broad refactor or a from-scratch idea implementation.

Clear boundaries

This skill owns exploratory code modifications only.
It must keep work isolated from the trusted baseline.
Use ai-research-explore instead when the task spans both current_research coordination and exploratory runs.
It may hand off execution to minimal-run-and-audit or run-train.
It should favor source-anchored copying and minimal adaptation over freeform rewrites.

Output expectations

explore_outputs/CHANGESET.md
explore_outputs/TOP_RUNS.md
explore_outputs/status.json

Notes

Use references/explore-policy.md, scripts/plan_code_changes.py, and scripts/write_outputs.py.