Discover and score Claude Code skills from public sources for database merging.
npx skills add https://github.com/somachak/claude-code-skills-db --skill hunter-somachak
Please help me install this Agent Skill. Skill: hunter Source: https://github.com/somachak/claude-code-skills-db/tree/main/hunter Command: npx skills add https://github.com/somachak/claude-code-skills-db --skill hunter-somachak
Daily pipeline that discovers Claude Code skills on the public web, scores them against a strict quality rubric, and merges eligible candidates into the skills database to keep the library fresh and consistent.
Run the hunter to update the skills database and commit changes to the repository.
High-intent search queries and answers about installing and using this skill.
Automated skill discovery is handled by a daily pipeline that searches GitHub, curated repos, RSS feeds, and the web to identify potential Claude Code skills for library expansion. It scores candidates against a quality rubric to determine eligibility before merging.
Skill deduplication works by scoring candidates against a quality rubric and dedup rules that decide whether to create, overwrite, or skip entries. Approved winners are merged into skills-database.json while logging all changes for audit.
Yes, the skill discovery pipeline applies a stack-aware rubric targeting React, Next.js, TypeScript, Node, Python FastAPI/Django, and Tailwind/shadcn to determine relevance and actionability for your specific technology stack.
The best way to keep a skills database fresh is running a daily pipeline that discovers candidates, scores them against a strict quality rubric, merges eligible winners into skills-database.json, and regenerates artifacts with logged changes for audit.
Yes, pyyaml is a required dependency for running the skill discovery workflow. It is needed to parse configuration and process the pipeline that identifies, scores, and merges Claude Code skills into the database.
The pipeline skips candidates that do not pass the stack-aware quality rubric or fail dedup rules. It evaluates relevance and actionability for targeted stacks like React, Next.js, Python, and Tailwind before deciding to create, overwrite, or skip.
emiled16Build automated agents that collect, enrich, and store public data from websites, APIs, and RSS feeds.
pddhktCapture session learnings and prepare commit-ready updates for the kotlin-kmp-skills repository.
hulusi-tuncProvide ready-made shadcn/ui component patterns for React projects.
dvy1987Audit all SKILL.md files and score them against a validation rubric.
