Trial
The biggest aggregator — 71,000-96,000+ skills scraped from GitHub. Discovery is the value prop.
Why It Matters
skillsmp.com aggregates skills from across GitHub, covering Claude, Codex, and ChatGPT ecosystems. At 71,000-96,000+ skills, the sheer scale makes it the widest discovery surface available. Think of it as a search engine for skills, not a curated store.
Strengths
- Largest skill catalog by far (71,000-96,000+) — unmatched for discovery
- Covers multiple agent platforms: Claude, Codex, ChatGPT
- Independent community project, not tied to a single vendor
Limitations
- Aggregation means wildly variable quality — many skills are untested or abandoned
- No security scanning or vetting; you inherit whatever GitHub has
- Discovery is useful but you still need to evaluate each skill yourself
- Quantity over quality: most of the catalog is noise you have to filter through
Risks
- 71,000-96,000+ skills is a vanity metric — the vast majority are forks, stubs, or abandoned experiments
- Scraping GitHub without vetting means malicious skills sit alongside legitimate ones; there's been no reported incident yet, but the attack surface is enormous
- No update tracking: a skill you install today could be compromised upstream tomorrow and you'd never know
- The "search engine for skills" pitch falls apart when search quality is poor — finding the right skill in 96K results is its own problem