explore-code
We need to translate the given English text into Hindi. The text describes a skill called "explore-code" but the instruction says to preserve the name only if it appears in the source text. The name "explore-code" does not appear in the provided text. So we just translate the text inside <text>. The text is a description of a skill for rigorous improvement implementation in deep learning research repositories. We need to preserve technical terms like "LoRA", "adapter layers", "backbone", "head", "rollback-aware records", "explore_outputs/", "current_research". Also preserve URLs if any (none here). Numbers: none. Protocol names: none. Product names: none. We translate into Hindi, keeping technical terms as is or with transliteration if needed. The instruction says "preserve" so likely keep them in English. But for Hindi translation, we can write them in English script within Hindi text. Also note: "Do not include the name unless it appears in the source text." So we don't add "explore-code".
npx skills add https://github.com/lllllllama/rigorpilot-skills --skill explore-codeexplore-code
Use this as the Rigor Improve implementation leaf skill. The installed slug
remains explore-code for compatibility.
Use the shared operating principles in
../../references/agent-operating-principles.md; this skill should guide
bounded candidate code work without over-prescribing implementation details.
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-exploreinstead when the task spans both current_research coordination and exploratory runs. - It may hand off execution to
minimal-run-and-auditorrun-train. - It should favor source-anchored copying and minimal adaptation over freeform rewrites.
- It should record why a candidate change is meaningful, how to roll it back, and why it remains a candidate rather than a verified contribution.
Output expectations
explore_outputs/CHANGESET.mdexplore_outputs/SCIENTIFIC_CHANGELOG.mdexplore_outputs/COMPARABILITY_REPORT.mdexplore_outputs/TOP_RUNS.mdexplore_outputs/status.json
Notes
Use references/explore-policy.md, ../../references/research-rigor-principles.md, scripts/plan_code_changes.py, and scripts/write_outputs.py.