AI Stupid Level

@aistupidlevel/mcp

An MCP server for AI Stupid Level (ASL). It lets Claude, Cursor, VS Code or any other MCP client read ASL's LLM leaderboard, score history, drift status, incidents and provider status. With a router key it can also ask the ASL Smart Router which model it would pick for a prompt, or send the prompt through the router.

Ask your assistant things like:

• "Which model is best for coding right now, according to AI Stupid Level?"
• "Has Claude Opus 5.5 got worse this week?"
• "Were there any provider incidents in the last few days?"
• "Which model would the Smart Router pick for this prompt, and why?"

It runs locally over stdio and needs Node.js 18 or later.

Keys

The tools use two different keys, and you only need the one for the tools you want.

Key	Used by	Where to get it
ASL_DATA_KEY (asl_live_…)	leaderboard, model, history, drift, incidents, providers, recommendations, global index	https://aistupidlevel.info/account/data-keys (free tier available)
ASL_ROUTER_KEY (aism_…)	preview_route, ask_router, router usage	https://aistupidlevel.info/router/keys, plus at least one provider key in your account

A tool whose key is missing does not fail the server. It answers with a message that says which key it needs and where to create it.

ASL_API_KEY is also accepted: it is used as the data key if it starts with asl_, and as the router key if it starts with aism_.

Free-tier limits

A free Data API key allows 10 requests a day and 1 a minute. Paid plans allow 10,000 a day or more (pricing). The free tier is meant for trying the API, so the server is careful with it:

• Every answer is cached in memory for 15 minutes (ASL_CACHE_TTL). Asking the same question again, or for another vendor or a shorter list from the same leaderboard, makes no request, and model names are looked up in the cached leaderboard.
• One tool call makes at most one Data API request. The exception is a model given by name before any leaderboard has been fetched, which takes one extra request to look up the model ids.
• Once the daily quota is used up, the server stops sending requests until 00:00 UTC. A request refused for quota still counts against it.
• If a fresh answer cannot be fetched, the server returns the last cached answer (up to 24 hours old) and says how old it is.
• Every answer ends with the number of requests left today.

On a free key, calling several different data tools within a minute hits the per-minute limit. The tool then says when to try again. If the wait is 15 seconds or less (ASL_MAX_WAIT), the server waits and retries once by itself.

The newest-run view (period: "latest") needs a paid key. Free keys read the 24-hour, 7-day and 30-day views.

preview_route and get_usage are not counted against your Smart Router plan. ask_router is a real routed request: the router calls the model with your own provider key, so the provider bills you and the request counts against your plan. Pass sandbox: true, or set ASL_SANDBOX=true, to get mock answers from the real routing at no cost.

Install

The package runs with npx, so there is nothing to install first. Put your keys in place of the placeholders, and leave out any key you do not have.

Claude Code

claude mcp add aistupidlevel -e ASL_DATA_KEY=asl_live_... -e ASL_ROUTER_KEY=aism_... -- npx -y @aistupidlevel/mcp

Add -s user to make it available in every project.

Claude Desktop

Settings → Developer → Edit Config, then add the server to claude_desktop_config.json:

{
  "mcpServers": {
    "aistupidlevel": {
      "command": "npx",
      "args": ["-y", "@aistupidlevel/mcp"],
      "env": {
        "ASL_DATA_KEY": "asl_live_...",
        "ASL_ROUTER_KEY": "aism_..."
      }
    }
  }
}

Restart Claude Desktop afterwards.

Cursor

Add the following to ~/.cursor/mcp.json (for all projects) or .cursor/mcp.json (for one project):

{
  "mcpServers": {
    "aistupidlevel": {
      "command": "npx",
      "args": ["-y", "@aistupidlevel/mcp"],
      "env": {
        "ASL_DATA_KEY": "asl_live_...",
        "ASL_ROUTER_KEY": "aism_..."
      }
    }
  }
}

VS Code

Add the following to .vscode/mcp.json (or to your user configuration with MCP: Open User Configuration). VS Code prompts for the keys and stores them securely:

{
  "inputs": [
    { "type": "promptString", "id": "asl-data-key", "description": "AI Stupid Level Data API key (asl_live_...)", "password": true },
    { "type": "promptString", "id": "asl-router-key", "description": "AI Stupid Level router key (aism_...)", "password": true }
  ],
  "servers": {
    "aistupidlevel": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@aistupidlevel/mcp"],
      "env": {
        "ASL_DATA_KEY": "${input:asl-data-key}",
        "ASL_ROUTER_KEY": "${input:asl-router-key}"
      }
    }
  }
}

Other clients

Any client that can start a stdio server can run it: the command is npx -y @aistupidlevel/mcp, and the keys go in the environment variables listed under Configuration.

Docker

The image builds the server from source. From the repository root:

docker build -f packages/mcp/Dockerfile -t aistupidlevel-mcp .
docker run -i --rm -e ASL_DATA_KEY=asl_live_... -e ASL_ROUTER_KEY=aism_... aistupidlevel-mcp

Tools

All tools except ask_router are read-only.

Tool	Key	What it returns
get_leaderboard	data	Tracked models ranked by combined, coding, reasoning or tool-use score, or by score per list-price dollar. Each row has the score, trend, status and stability, and the model ids the other tools take. Filter by vendor; choose the view (24h, 7d, 1m, or latest on a paid key).
get_model	data	One model's details: vendor, version, notes, tool-calling support and days tracked.
get_model_history	data	One model's score for each run in the window, with a summary and the per-axis results of the newest run.
get_drift_status	data	One model's drift signature: regime, current score against its baseline, the change-point detector for each suite, the last significant change, and which axes moved.
get_incidents	data	Recent incidents (provider-wide and per model) with severity, the models affected, when each was detected and when it was resolved.
get_provider_status	data	Each provider's API status from ASL's health probes, with the probe's response time and the last error.
get_recommendations	data	ASL's current picks (best for coding, most reliable, fastest, best value, poor value, avoid now), each with its reason and the evidence behind it.
get_global_index	data	ASL's global index across the tracked models, now and over the last 24 hours.
preview_route	router	The model the Smart Router would pick for a prompt, its score and estimated cost, the reason, and the fallback chain. Nothing is called or counted.
ask_router	router	Sends a prompt through the Smart Router and returns the answer, the model and provider that answered, why it was chosen, the latency and the tokens used. Set sandbox: true for a mock answer at no cost.
get_usage	either	This month's router usage and budget, and the Data API requests left today.

Models can be given by numeric id, by name (claude-opus-5-5) or by display name (Claude Opus 5.5).

preview_route and ask_router take a routing mode (auto, auto-best, auto-coding, auto-reasoning, auto-tooling, auto-task, auto-value, auto-value-coding, auto-value-reasoning, auto-value-tooling, auto-fastest-quality, auto-consistent, auto-fastest, auto-cheapest), or pin_model to name a model directly. The router documentation explains each mode.

The server also provides one prompt, choose_model, which walks the assistant through choosing a model for a task you describe.

Answers from the Data API end with a source line. The data is licensed with attribution required: cite aistupidlevel.info when you republish it.

Configuration

Variable	Default	Meaning
ASL_DATA_KEY	none	Data API key (asl_live_…)
ASL_ROUTER_KEY	none	Smart Router key (aism_…)
ASL_SANDBOX	false	true puts every Smart Router call in sandbox mode: real routing, mock answer, nothing spent or counted
ASL_CACHE_TTL	900	Seconds a Data API answer is reused
ASL_MAX_WAIT	15	When the per-minute limit asks for a wait this short (seconds) or shorter, wait and retry once. 0 turns this off
ASL_TIMEOUT_MS	300000	Timeout for a Smart Router request

Keys are never logged. The server writes its diagnostics to stderr, because stdout carries the MCP protocol.

Use as a library

import { createServer } from '@aistupidlevel/mcp';
import { StdioServerTransport } from '@modelcontextprotocol/sdk/server/stdio.js';

const server = createServer({ dataKey: process.env.ASL_DATA_KEY, cacheTtlSeconds: 600 });
await server.connect(new StdioServerTransport());

createServer accepts the same settings as the environment variables, plus a custom fetch.

Development

npm ci && npm run build && npm test

The unit tests start the server in-process, connect an MCP client to it over an in-memory transport, and replay real API responses from test/fixtures. The live tests are skipped unless keys are set:

ASL_ROUTER_KEY=aism_... npm test    # Smart Router, sandbox only: nothing is spent or counted
ASL_DATA_KEY=asl_live_... npm test  # Data API: one request

Links

• Data API documentation
• Smart Router documentation
• TypeScript SDK for the router
• Leaderboard

MIT licence.