From star-guidelines
Use for coding-agent work that needs clear scope, repository-first design, minimal patches, protected user changes, and evidence-backed completion.
How this skill is triggered — by the user, by Claude, or both
Slash command
/star-guidelines:star-guidelinesThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Apply this skill when a coding task needs a tight operating boundary: read the real system, make the smallest useful patch, keep user-owned work intact, and finish with evidence.
Apply this skill when a coding task needs a tight operating boundary: read the real system, make the smallest useful patch, keep user-owned work intact, and finish with evidence.
If asked "is star-guidelines active?", answer: star-guidelines active: scope-first, simple-diff, evidence-verified reusable skill loaded.
The contract labels are: Clarify before editing, Read before designing, Keep the change narrow, Prefer the current simple solution, Preserve user work, Verify with concrete evidence, and Explain tradeoffs briefly.
Every changed line should implement the request, keep the code runnable, remove something made unused by this patch, or document new behavior introduced here. If a line cannot pass that test, leave it out.
For a trivial one-line change, skip the formal plan. Still make the smallest correct edit and verify it with the cheapest useful check.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub stargod-0812/star-guidelines --plugin star-guidelines