mcp-checkup
MCP server that analyzes your MCP setup: token costs per server/tool, grades, duplicate detection, optimization report
Documentation
mcp-checkup
Your MCP servers are eating your context window. Find out how much.
mcp-checkup is an MCP server that analyzes your MCP setup — measuring the token cost of every installed server and tool, finding duplicates, grading efficiency, and generating optimization reports.
Quick Start
npx mcp-checkupThen ask your AI assistant:
"Run an MCP health check on my setup"
"How many tokens are my MCP servers using?"
"Generate an MCP health report"Why?
Every MCP tool costs 550–1,400 tokens just for its schema. A server with 50 tools can eat 30,000+ tokens before you type anything. Most developers have no idea how much context window their MCP servers are consuming.
Real examples from the community:
- Adding GitHub MCP increased one user's token usage from 34K to 80K (source)
- One developer found 67,000 tokens consumed by just 4 servers before any conversation (source)
- Context7 (240K weekly downloads) has tool descriptions costing 3x more tokens than necessary (source)
Tools
`analyze_servers`
Scan your MCP config and measure the token cost of every installed server. Auto-detects config files.
"Which of my MCP servers costs the most tokens?"`analyze_tools`
Deep-dive into a specific server — per-tool token costs, grades, bloated descriptions, and optimization suggestions.
"Break down the token cost of my GitHub MCP server"`find_duplicates`
Find duplicate tool names across your installed servers. Redundant tools waste context for no benefit.
"Do any of my MCP servers have overlapping tools?"`count_tokens`
Count tokens in any text. Useful for estimating context usage.
"How many tokens is this prompt?"`generate_report`
Full markdown health report: server grades (A-F), token costs, duplicates, and actionable recommendations.
"Generate a full MCP health report"Grading System
Per-Tool Grades
| Grade | Tokens | Meaning |
|---|---|---|
| A | ≤100 | Excellent — minimal context cost |
| B | ≤300 | Good — reasonable schema size |
| C | ≤600 | Fair — could be optimized |
| D | ≤1000 | Poor — bloated schema |
| F | >1000 | Failing — severely impacts context window |
Per-Server Grades
| Grade | Total Tokens | Meaning |
|---|---|---|
| A | ≤500 | Lean and efficient |
| B | ≤1500 | Reasonable |
| C | ≤3000 | Getting heavy |
| D | ≤6000 | Significant context cost |
| F | >6000 | Major context window drain |
Installation
Claude Desktop / Claude Code
{
"mcpServers": {
"mcp-checkup": {
"command": "npx",
"args": ["-y", "mcp-checkup"]
}
}
}Cursor
Add to `.cursor/mcp.json` with the same format.
Auto-Detection
mcp-checkup automatically finds your MCP config in these locations:
- `.mcp.json` (current directory)
- Claude Desktop config (Windows/macOS/Linux)
- `.cursor/mcp.json`
Or pass a custom path to any tool.
How It Compares
| Tool | Purpose | Approach |
|---|---|---|
| mcp-checkup | Diagnose token costs | Analyze & report — see exactly what each tool costs |
| lean-ctx | Reduce token usage | Compress context at runtime |
| MCP Inspector | Debug MCP servers | Test connections and tool calls |
Use mcp-checkup first to find the problem, then decide how to fix it.
Part of the MCP Toolkit
- webcheck-mcp — Website health analysis
- git-summary-mcp — Git repository intelligence
- mcp-checkup — MCP setup health analyzer
- dev-utils-mcp — Developer utilities
- codescan-mcp — Codebase health scanner
- deadlink-checker-mcp — Dead link detector
License
MIT
Frequently asked questions
What is mcp-checkup?
mcp-checkup is MCP server that analyzes your MCP setup: token costs per server/tool, grades, duplicate detection, optimization report
How do I install mcp-checkup?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is mcp-checkup open source?
Yes — it is hosted on GitHub at https://github.com/yifanyifan897645/mcp-checkup and has 1 stars.
Related MCP tools
Cut AI token costs 95%+ on code exploration. The leading MCP server for precise, symbol-level GitHub code retrieval via tree-sitter AST. Works with Claude Code, Cursor & any MCP client. 313B+ tokens saved.
AI Skills, MCP Tools, and CLI for Unity Engine. Full AI develop and test loop. Use cli for quick setup. Efficient token usage, advanced tools. Any C# method may be turned into a tool by a single line. Works with Claude Code, Gemini, Copilot, Cursor and any other absolutely for free.
Official remote MCP server for Atlassian. Securely connect Jira, Confluence, Jira Service Management, Bitbucket, and Compass to Claude, ChatGPT, Cursor, VS Code, and other AI tools using OAuth 2.1 or API tokens.
Open-source coding agent memory. Records issues, attempts, fixes and decisions, then warns your agent before it repeats an approach that already failed. Native MCP server for Claude Code, Cursor, Antigravity and Codex. 100% local, no cloud, no telemetry. MIT.
The go-to web for your AI coding agent — local-first search, fetch, crawl & research over MCP. No API keys, no cloud, $0/query. Public beta.
Universal AI context generator. Saves thousands of tokens per conversation in Claude Code, Cursor, Copilot, Codex, and more.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP