What Skill Store 33 is
Skill Store 33 is a catalog of agent skills: SKILL.md folders that tell an AI agent such as Claude Code, Cursor or Codex how to carry out a specific task. It holds more than 61,000 of them, across software, data, security, business, finance, legal, health, creative work and more.
You can browse and download skills on the site, or let your agent search the store itself through our MCP server. Your agent describes the task, gets a few matching skills back, and loads the one that fits.
Who runs it
Skill Store 33 is run by ReadyAI. The legal entity named in our Terms and Privacy Policy is Afterparty, Inc., doing business as Skill Store 33.
Where the skills come from
Almost every skill in the store is imported from public sources, mainly the skills.sh index of public GitHub repositories. We test skills in our sandbox: more than 200 have passed so far and show as Verified.
A small set, around a hundred, comes from vendor and community repositories we track directly, such as skills published by the company whose product they drive.
A third source is SN33, ReadyAI's subnet on the Bittensor network. ReadyAI has designed a skill task for the subnet: miners write a skill together with a test plan of concrete input and output checks, and validators score how specific and correct those checks are and how much of the requested skill they cover. So far, 13 skills in the store come from SN33. The store does not treat subnet scores as a quality rating; skills from SN33 go through the same checks as any other.
What each listing shows
What the skill covers.
Its publisher, and its source: the upstream repository it came from.
Its licence, where we know it.
Its verification state.
An approximate size in tokens, so you can see how much of your agent's context it uses.
The SKILL.md to download. When a skill's licence doesn't allow us to redistribute it, the store points you to the author's copy instead.
Verification states
Not yet verified: listed, but not tested by us yet.
Safety scanned: the skill passed a safety scan. Its behaviour has not been tested.
Verified: the skill's own tests ran in our sandbox and passed.
Tests failing: the skill's tests ran in our sandbox and did not pass.
Whatever the state, read a skill before your agent runs it. A skill can tell an agent to run code, change files or send data elsewhere.
Contact us
Email support@readyai.ai for questions, problems with a skill, or to have your own skill's listing corrected or removed. ReadyAI is on X as @ReadyAI_.