2 findingsStatic scanned
MCP server that serves knowledge about your components based on your Storybook stories and documentation
9.4M installs/moMIT
Rubric v1.0 · 16 checks, published in full
Knowwhatyou'reinstalling.
987 of 990 read against 16 published checks, with what was found printed on the card.
None has been through the full rubric yet. Every card says which of the 16 published checks it has and has not been put through.
Storybook MCP· storybookChrome Devtools MCP· mathiasContext7 MCP· upstashServer Filesystem· modelcontextprotocolPandacss MCP· pandacssServer Pdf· modelcontextprotocolYahoo Finance2· gadiccNotion MCP Server· notionhqAgent Device· thymikeeServer Sequential Thinking· modelcontextprotocol
Start with what you're after
Four kinds of thing, one standard behind all of them.
What reading 987 of them turned up
Not failures. Things worth knowing before you hand something to an agent, each one printed in full on the listing it came from.
Read the full audit, with its method, limits and raw data0
listings listed
0
makers with a storefront
0
passing all 16 checks
0%
kept by founding makers
Founding makers keep 100% for their first 6 months. After that, 10% on what a maker ships themselves and 15% when we host it. Buyers pay 0% on top.
New way to browse
See the whole market in 3D
Stand inside a sphere of every listing and turn the room around you. Drag to look, scroll to travel, click a card to open it. It is the same catalogue, one click away.
Most wanted right now
What people are asking for the most. The only row here ranked by demand.
Browse by what it does
Verified
A certification whose criteria are private is an opinion. Ours are published and versioned, so anybody can check what the badge is worth, including the parts it is not worth.
Covers
- No credentials in shipped code or repository history
- No known-vulnerable dependencies on a reachable path
- Declared capabilities match observed behaviour
- Declared list of hosts it may contact
Does not cover
- Whether it is any good at your particular job
- What the maker does with data after it leaves your systems
- Whether the maker will still be maintaining it next year
- Whether using it is lawful in your industry or jurisdiction
Free, and staying free
Our library
86 guides, definitions, calculators and checklists on judging software you did not write. No account needed.
Guides 16
How to judge an agent in ten minutes, read a findings report, and refuse the right permissions.
Glossary 60
Every term that made you nod along, each one naming the mistake people make with it.
Calculators 3
What metered pricing really bills, build versus buy, and the blast radius of a permission set.
Free downloads 7
Checklists to print, cheat sheets that translate the jargon, guides to send to a colleague.
Cannot find it? Ask.
Requests are public asks for software nobody has built here yet, ranked by how many people want it. Every entry is somebody else's market research, done for free and published, and it is where most of what gets built next comes from.
See what people wantSelling here
Payouts are held 14 days, then released. Three things fail most first attempts: a secret left in repository history, a manifest that does not match the software, and no declared egress.
Start selling