Where does the MCP catalog come from, and how fresh is it?

Undocumented. The README says only that a goal is matched against a catalog of MCP servers or capabilities — no source, no format, no refresh story, and no statement about whether servers are discovered from a registry or shipped with the repo.

What does two-stage when large actually split up?

Only that a two-stage path exists for large catalogs is stated. Which stage does retrieval and which does the Jev call, and what the second stage is given, are not described, so the architecture cannot be judged from the README.

How does it choose between two servers that both plausibly fit?

Not described. No tie-break rule, no confidence threshold and no abstain path is mentioned, which makes a confident wrong match a plausible failure mode with no stated guard against it.

Can I try it without installing any MCP servers locally?

The README gives npm i and npm run dev and says nothing about a registry, a key or a working directory of servers. Whether it runs standalone against a static catalog is not stated, so assume you may need to supply one.

What are the setup commands for McpMatch?

npm i then npm run dev. The README gives no key, config file, or environment variable beyond those commands.

What language is McpMatch written in?

It is cataloged as TypeScript. The README names no framework, runtime version, or dependencies.

Is McpMatch part of a larger set of Jev demos?

Yes. It is one of eight demos cataloged under the GitHub user ndolinschi, alongside JevPlay, LaneBreak, PulseLane, SpendBrake, SwarmRouter, ToolGate, and TrustGate.

CommunityJev Demos & Experiments

Mcpmatch

McpMatch takes a free-text user goal and matches it against a catalog of MCP servers or capabilities using TypeSafe Jev, with a two-stage path for large catalogs. Runs with npm i and npm run dev.

See highlights

Open github.com

Stars: 0
Language: TypeScript
Published by: Community
Author: ndolinschi
Last verified: Oct 2026

ndolinschi/mcpmatch

User repository on GitHub

View on GitHub

Match user goals to MCP catalog (two-stage) via TypeSafe Jev

Forks: 0
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

About this project

User goal → match the best MCP server/capability from a catalog (two-stage when large) via TypeSafe Jev.

Quoted from the project's README. Read the full README.

Tagscommunity jevbest catalog-import awesomejev

Works forJev for Classification Jev for Tool calling

Used inJev for Software & DevTools

Languages: TypeScript

Highlights

Input is a free-text user goal; output is a match from a catalog of servers or capabilities.
A two-stage matching path is used when the catalog is large, per the README.
Setup is npm i && npm run dev; catalog source and key handling are undocumented.

Watch out

The README is one sentence plus install commands. It does not say where the catalog comes from or how it is loaded, what best means when several servers fit, how the two stages split the work or what the second stage receives, whether the catalog is bundled or fetched live, or whether an API key is required. There is no evaluation against a labelled set of goals, no match accuracy and no latency or cost figure, so match quality is unmeasured. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

jevarena

Next.js spectator demo in which two independent Jev agents drive separate Snake browsers through same-origin iframes, clicking indexed direction buttons only, with each board state streamed to a server-only endpoint that returns the chosen target.