mcp-slim
Optimizes MCP tool menus from OpenAPI specs by grouping operations by tag and trimming schemas to reduce token usage, improving agent tool-selection accuracy.
README
mcp-doctor ? Agent Readiness Platform (MCP wedge)
Agent-facing API QA ? score, inspect, evaluate, and improve MCP quality.
"We prove agents can actually use your MCP."
Real benchmark results
State of MCP Quality 2026 (v0) ? 10 public servers scored live:
| Server | Grade | Tools | Tokens |
|---|---|---|---|
| MCP Filesystem | A | 14 | 1,997 |
| MCP Sequential Thinking | A | 1 | 996 |
| MCP Everything | A | 13 | 1,236 |
| MCP Memory | A | 9 | 1,040 |
| MCP Puppeteer | B | 7 | 612 |
Per-server reports: examples/reports/
npx github:louisreid/mcp-doctor benchmark -o ./reports
Three pillars
| Pillar | Command | Status |
|---|---|---|
| Static scorecard | inspect, test |
v0.4 |
| Task success | eval (BYOK) |
v0.4 |
| Agent friction | included in eval |
v0.4 |
Plus: Recommended Improvements, Replay Timeline, Model Compatibility Matrix (multi-model eval).
Quick start
# Benchmark public MCPs
npx github:louisreid/mcp-doctor benchmark
# Inspect your Cursor MCP server
npx github:louisreid/mcp-doctor list
npx github:louisreid/mcp-doctor inspect <name> -o report.md
# Agent eval (BYOK � Vercel AI Gateway free trial, local only)
export AI_GATEWAY_API_KEY=...
npx github:louisreid/mcp-doctor eval memory \
--task "List all tools and describe them" \
--models openai/gpt-4o-mini,openai/gpt-4o -o eval-report.md
AI Gateway setup (one-time)
vercel login
cd mcp-doctor && vercel link
vercel ai-gateway api-keys create --name mcp-doctor-local --budget 5 --refresh-period monthly
# Key saved to .env.local (gitignored) � CLI auto-loads it for eval
Pain Interview (before sending to friends)
See docs/FRIEND-GUIDE.md and Coefficient PAIN-INTERVIEW.md.
Do not ask friends to beta-test until you've run a 30-min workflow interview.
Commands
| Command | Description |
|---|---|
benchmark |
State of MCP Quality ? score catalog servers |
list |
MCP servers in ~/.cursor/mcp.json |
inspect <name> |
Live connect + scorecard + suggested fixes |
eval <name> --task "..." |
BYOK agent eval + friction + replay (Vercel AI Gateway) |
test --demo |
Static scorecard on OpenAPI fixture |
competitors |
36-player market map |
Roadmap
| Priority | Feature | Status |
|---|---|---|
| 1 | Public benchmark reports | v0.4 |
| 2 | Pain Interview (Jonty / Anders) | In progress |
| 3 | BYOK eval + friction | v0.4 |
| 4 | Suggested fixes | v0.4 |
| 5 | Model matrix | v0.4 |
| 6 | Scale to 50 MCPs + awards | Next |
| 9 | OpenAPI drift | Deprioritized |
License
MIT ? Coefficient investigation
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.