AI assistants and search decision
Agentskill
A small, searchable skills directory plus a CLI installer is realistic for one developer to build and run, but reproducing the vendor's large catalog, curated quality/security audits, and marketplace ecosystem is not practical without the original dataset and community.
Visit website↗Not priced
No pricing page we fetched carried a figure, so there is nothing to compare against. The build side is still real.
$100one-off52 h to build
$50/mo6 h/mo upkeep
No published price to break even against.
Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need - the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship. All Agentskill alternatives, with the arithmetic →
What a replacement has to do
- Search and discover skills → view skill metadata and ratings → install a skill via CLI (npm/x) → serve skill metadata and search API
What it still won’t have
- The existing catalogue scale (275,000+ indexed skills)
- Curated quality and security audit scores across hundreds of thousands of skills
- Marketplace ecosystem of creators, bundles, and reviews
- Polished UX and CI for continuous submission and automated audits
What remains hard
- Proprietary data
First-year cost
No published price
Agentskill does not publish a price we could read, so there is nothing to compare against. What building costs is below.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal AI-agent-skills directory and installer using Next.js (React) frontend, a Node.js + Express REST API, SQLite (or Postgres) for metadata, and Lunr.js for search indexing. Core features: (1) import script to ingest a JSON/CSV skill catalog into the DB; (2) REST endpoints for search, listing, and skill detail; (3) React UI with search, filters, and skill detail pages showing metadata and scores; (4) an npm CLI (npx) that fetches a skill manifest and installs or links the skill locally; (5) deployment scripts for Vercel (frontend) and a small cloud VM or managed DB for the API; (6) simple authentication for creator submissions (optional). Out of scope: building a large-scale catalog or automated security audits. Include error handling, basic tests (unit for API routes and integration for import + CLI), and README deployment instructions.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 4 cited sources+3
- Hard moats found in the evidence-3
- Evidence score57
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 4
Every page the run actually retrieved.
- official productAgentskill homepage
- official productAgentskill product pages / for/product
- open sourcehigress-group/himarket
- open sourcearc53/DocsGPT
Integrity checks
What held up, and what did not.
