MCP.Market

ModelShortlist MCP server

by AgenticArtists·io.github.AgenticArtists/modelshortlist·v0.2.4

Local BYOK MCP for AI model shortlisting using OpenRouter + Artificial Analysis.

B83/100grade B
What users say
No reviews yet
Be the first
Safety scan
B83/100

full report

Adoption
Growing

0 stars54 downloads/wk

Reviews

Write one

Nobody has reviewed ModelShortlist yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

ModelShortlist tools (3)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • compare_models

    Use this after model discovery for a focused comparison of specific OpenRouter model IDs. It returns OpenRouter catalog data, ZDR availability, Artificial Analysis metrics when confidently matched, and explicit source freshness metadata. ZDR is not assumed to be required.

  • modelshortlist_status

    Use this to diagnose source freshness, OpenRouter catalog coverage, model matching coverage, ZDR availability, or Artificial Analysis quota state.

  • recommend_models

    Use this whenever the user asks which AI model to use for a workload. It considers the OpenRouter model catalog and adds Artificial Analysis benchmark data when confidently matched, while returning explicit source freshness metadata if a refresh is degraded. Zero Data Retention is optional and must only be enforced when the user explicitly requires ZDR. After calling this tool, YOU must make the r

Public scan report

scanner v0.1.10 · 2026-10-03 · same rubric, same numbers if you re-run it

no findings
  • Code scan9 source files scanned25/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitystatic API keys via environment variables6/15
  • Maintenancelast push 14 days ago15/15
  • Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
Overall 83/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How ModelShortlist compares

Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
83/100
higher than 68% of 4,460 AI and agents servers
Tools
3
the typical AI and agents server has 6
Tools that change things
0%
typical: 0%. Fewer write tools means less an agent can break
Downloads a week
54
more than 45% of 1,538 AI and agents servers with a package

What its tools let an agent do: 3 read

Install ModelShortlist in Claude Code, Cursor or VS Code

Runs npx -y @agentic.artists/modelshortlist on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add modelshortlist -- npx -y @agentic.artists/modelshortlist
Add to Cursor

What the publisher says

From the ModelShortlist repository's README, as published. We do not edit it. Read it on GitHub

ModelShortlist

Stop guessing which AI model to use.

ModelShortlist is a local, bring-your-own-key MCP server that gives your AI assistant model-selection evidence from the full OpenRouter model catalog plus Artificial Analysis benchmarks. It refreshes upstream evidence on demand/cache expiry and explicitly reports when any source is stale or unavailable. Zero Data Retention (ZDR) is an optional hard constraint only when you explicitly require it.

No hosted ModelShortlist backend. No account. No telemetry in the MCP. Your API keys stay with the local MCP process and are used to call the upstream services directly.

  • Website: modelshortlist.com
  • Fast install configurator: modelshortlist.com/install
  • Model-selection guides: modelshortlist.com/guides
  • npm: @agentic.artists/modelshortlist
  • Official MCP Registry: io.github.AgenticArtists/modelshortlist
  • License: MIT

Why ModelShortlist

Model choice is no longer just "which model has the highest benchmark score?" The right answer depends on the workload, capabilities, cost, context, availability, and privacy requirements you actually have.

ModelShortlist helps your chat agent reason over:

  • the OpenRouter model catalog
  • tool/function-calling support
  • context and completion limits
  • OpenRouter input/output pricing, including tiered pricing when OpenRouter publishes it
  • ZDR endpoint availability when privacy requires it
  • ZDR endpoint latency, throughput, uptime, and provider options when applicable
  • Artificial Analysis Intelligence Index
  • Artificial Analysis Coding Index
  • Artificial Analysis Agentic Index
  • Artificial Analysis pricing and median performance
  • explicit source freshness and degraded-state warnings

By default, ZDR is not an eligibility requirement. ModelShortlist considers the full OpenRouter catalog. If you explicitly require ZDR, it switches to OpenRouter ZDR endpoint evidence and requires hard constraints to be satisfied by the same real endpoint.

The host AI makes the final recommendation based on your use case. ModelShortlist deliberately does not impose one universal ranking formula.

Fastest install

Use the browser-only configurator:

Install ModelShortlist →

It generates client-specific config or commands for:

  • Claude Desktop
  • Hermes Desktop
  • Cursor
  • Claude Code
  • VS Code / Copilot

You paste your own Artificial Analysis and OpenRouter keys into the configurator. They are used in your browser to generate configuration text and are not sent to ModelShortlist or a ModelShortlist backend.

Requirements:

  • Node.js 20+
  • an Artificial Analysis API key
  • an OpenRouter API key
  • an MCP-capable client

Generic npm config

Most local stdio MCP clients, including Claude Desktop, Hermes Desktop, and Cursor, can launch ModelShortlist directly with npx:

{
  "mcpServers": {
    "modelshortlist": {
      "command": "npx",
      "args": ["-y", "@agentic.artists/modelshortlist"],
      "env": {
        "ARTIFICIAL_ANALYSIS_API_KEY": "YOUR_KEY",
        "OPENROUTER_API_KEY": "YOUR_KEY"
      }
    }
  }
}

On Windows GUI clients where npx is not available on the app's PATH, use npx.cmd or use the guided clone setup below.

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

ModelShortlist: common questions

Is ModelShortlist MCP server safe?
Mostly: it is graded B (83/100). Read the ModelShortlist safety report
How do I install ModelShortlist?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does ModelShortlist need an API key?
Yes. The registry entry asks for ARTIFICIAL_ANALYSIS_API_KEY, OPENROUTER_API_KEY.
Is ModelShortlist maintained?
The last commit was 46 days ago (2026-09-19). The latest release is v0.2.4.
What can I use instead of ModelShortlist?
Servers from other publishers that do the same job: MCP Server for WinDbg Crash Analysis, Stata MCP server and Depwire MCP server.
How does ModelShortlist compare with other AI and agents MCP servers?
safety score 83/100 (higher than 68% of 4,460 AI and agents servers); tools 3 (the typical AI and agents server has 6); tools that change things 0% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.

Alternatives to ModelShortlist

Same job from other publishers: the closest match first, then the best rated.

  • MCP Server for WinDbg Crash Analysis
    A Model Context Protocol server for Windows crash dump analysis using WinDbg/CDB
    A
  • Stata
    Let LLM help you achieve your regression analysis with Stata
    C
  • Depwire
    Dependency graph + 24 MCP tools. Impact analysis, simulation, security, agent coordination
    B