Cross-agent skill registry that's actually curated, not scraped
There are now 4+ competing 'npm for AI agent skills' registries (Skills.sh, SkillsMP, ClaudeSkills.info, Agensi, awesome-agent-skills) and they mostly index by crawling GitHub for SKILL.md files. Devs running Claude Code, Codex CLI, Cursor, and Gemini CLI simultaneously want one trusted source where skills are tested against multiple agents, version-pinned, and not malware. Demand is for a curated layer over the scraped chaos, not yet another scraper.
The defensible play is the test matrix, not the catalog. Anyone can scrape SKILL.md files. Almost nobody is paying the compute bill to actually run each skill against four agents on every release and publish the pass/fail.
landscape (4 existing solutions)
The category split: scrapers compete on volume, curators compete on trust. The under-served niche is 'I run three agents and want one skill that works in all of them with proof.' A small CI matrix that runs each submitted skill against the four major agents would be a moat.