skills
A skill is executable trust — you install someone's instructions and their scripts into an agent that can touch your files, your keys, your machine.
A supply chain needs a gate. This is a curated catalog of skills reviewed for what they contain, how they behave, and what they ask an agent to access.
boundary
audit
by design
Categories
10 categories · 46 skillssecurity
Score any skill, repo, or agent answer 0–8 with file:line evidence, before you trust it.
token-efficiency
Make long-running agents spend attention deliberately.
rag
Ground an answer in sources an operator can inspect.
memory
Keep durable context explicit, linked, and removable.
apple
Useful local actions with a visible permission surface.
swift
Swift beyond the app: embedded, Android, Vapor, ports.
workflow
Turn a vague task into inspectable steps.
tools-services
Connect agents to services without hiding the blast radius.
web
SEO without tricks, and wallet integration on the web.
iot
Small interfaces deserve the same scrutiny as large ones.
The auditors inspect manifests, instructions, permissions, and scripts. They never execute the target skill during review; the score is a review signal, not a promise of safety.