Legitshow Search

io.github.commitshowPublished by GitHub account commitshowv0.1.0Updated Oct 24, 2026

Search launched software (SaaS, AI tools, MCP servers) by measured production-readiness.

VerifiedSTDIODesktop onlyWeb Search & ScrapingDeveloper Tools

Overview

AI-generated overview

A local stdio MCP server that lets an assistant search Legit.Show's directory of launched software and filter results by measured production-readiness scores.

What you can do
Search launched software (SaaS, AI tools, MCP servers) by name, category, open-source status and minimum per-frame scores, returning each match with its scores and page link. Retrieve the canonical Legit.Show documentation, including the 7-Frame benchmark and how to cite a service. Results are limited to per-frame scores, a summary and the page; no combined overall score is provided.
When to use it
Useful when you want an assistant to recommend tools based on measured production-readiness (performance, security, privacy and similar) rather than popularity or GitHub stars, for example finding open-source MCP servers with a security score above a threshold.
Requirements
Runs locally via npx (Node.js) as a stdio process. No API key or account is needed. It calls the public Legit.Show REST API over the network; the endpoint can be overridden with the LEGIT_SEARCH_API and LEGIT_DOCS_BASE environment variables.
Before you install
The README states this server has been merged into commitshow-mcp and that both the npm package and the registry entry are deprecated, so install the successor instead. Queries and filters are sent to the third-party Legit.Show API. Scores reflect only the public surface measured by the provider and are not an overall rating.

Installation

In SourceWeft

  1. Open Legitshow Search in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

[!IMPORTANT] This server has been merged into commitshow-mcp and is no longer published.

commitshow-mcp carries the same two tools — search_services and fetch_docs, with identical arguments — and adds audit_repo and project_status. Two entries under one name, half of them the same, gave anyone browsing the registry no way to tell which to install; there is one now.

jsonc
{ "mcpServers": { "legitshow": { "command": "npx", "args": ["-y", "commitshow-mcp@latest"] } } }

The npm package and the registry entry io.github.commitshow/legitshow-search are both marked deprecated. The source stays here as the record of what it was.

Legit.Show MCP server

Model Context Protocol server for Legit.Show. Plug it into Claude Desktop, Cursor, Cline, Windsurf, or any MCP host, and your model can search a directory of launched software by measured production-readiness — not GitHub stars — without ever leaving the chat.

[npm] [MIT license] [node 20+]

jsonc
// any MCP host config (Claude Desktop · Cursor · Cline · Windsurf · …){  "mcpServers": {    "legitshow": {      "command": "npx",      "args": ["-y", "legitshow-mcp"]    }  }}

That's the entire install. No API key. The server is a stdio shim over the public https://legit.show/api/search REST surface — one source of truth for the data.

Why this is different

Most "find me a tool" flows rank by GitHub stars or popularity. Legit.Show carries a deterministic 7-Frame production-readiness benchmark (Performance, Accessibility, Security, Privacy, Reliability, Standards, Discoverability) measured from the public surface — real Lighthouse, HTTP headers, live probes, no LLM in the scoring path. So your agent can filter by whether something is actually production-ready:

"find production-ready, open-source MCP servers with Security ≥ 80"

Tools exposed

ToolWhat it does
search_services({ query?, category?, min_scores?, is_open_source?, limit? })Search launched software by name or filter. Returns each match with its per-frame scores and its legit.show/s/<slug> page.
fetch_docs()Pull the canonical Legit.Show docs (llms.txt) — what Legit.Show is, the 7-Frame benchmark, and how to cite a service.

Boundary: per-frame scores + summary + the page only. There is no combined overall total in the public data (it's a licensed product), so the model won't invent one.

Host wiring snippets

Claude Desktop — claude_desktop_config.json

jsonc
{ "mcpServers": { "legitshow": { "command": "npx", "args": ["-y", "legitshow-mcp"] } } }

Cursor — .cursor/mcp.json

jsonc
{ "mcpServers": { "legitshow": { "command": "npx", "args": ["-y", "legitshow-mcp"] } } }

Cline / Continue / Windsurf

Same three-line npx -y legitshow-mcp invocation in the host's MCP config.

No MCP host? Plain REST

GET https://legit.show/api/search?q=<name>

Filters: category, min_scores (e.g. {"security":80}), is_open_source, limit (≤50). GET or POST. No key.

Self-hosting / dev overrides

EnvDefaultUsed by
LEGIT_SEARCH_APIhttps://legit.show/api/searchsearch_services
LEGIT_DOCS_BASEhttps://legit.showfetch_docs (llms.txt)

Links

  • Legit.Show — the directory: every launched service, tested
  • Methodology — how the 7-Frame benchmark is measured
  • llms.txt — the canonical machine-readable doc

MIT · operated by Madeflo, Inc.

Source: legitshow-mcp/README.md at commit bcdb9dd

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.1.0LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts