Auto research loop
Optional report-only self-improvement loop that studies approved public sources, classifies improvement ideas, and preserves human approval gates before any behavior changes.From its SKILL.md
npx -y skills add GunsNR/smart-skill-user --skill auto-research-loopAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.5 KB, 268 tokens by cl100k_base, as published. Nobody here has run it
Auto-Research Loop
Use this skill when a maintainer explicitly asks for optional auto-research or self-improvement reports.
Scope
This skill is for report generation and idea triage only. It does not authorize commits, pull requests, releases, publishing, new integrations, or copied external code.
Workflow
- Run Smart Skill Preflight first and confirm the target repository.
- Read
config/research-sources.yml. - Prefer offline or cached runs unless the maintainer asks for bounded online research.
- Generate a report under
research/. - Classify ideas by impact, risk, effort, source, and license sensitivity.
- Ask for explicit maintainer approval before implementing any idea that changes behavior.
Safety Rules
- Use only approved public sources.
- Do not clone whole external repositories by default.
- Do not copy external code or distinctive examples without license review.
- Do not include private client data, local paths, secrets, screenshots, or videos in reports.
- Do not create issues, pull requests, releases, or published packages by default.
- Treat network failures as report warnings, not normal validation failures.
Local Command
python scripts/auto_research.py --offline --dry-run --output research/auto-research-latest.md
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.