toolrank

The first ATO (Agent Tool Optimization) platform. Score and optimize MCP tools so AI agents choose yours.

1 stars AstroOthers Updated Jun 15, 2026

Documentation

We scanned 4,162 MCP servers. Here's what we found.

Metric	Value
Registered servers	4,162
With tool definitions	1,122 (27%)
Invisible to agents	3,040 (73%)
Average score	84.7/100
Selection advantage	3.6x for optimized tools

> 73% of MCP servers are invisible to AI agents. They have no tool definitions, no descriptions, no schema. When an agent searches for tools, these servers don't exist.

Sources: arXiv 2602.14878, arXiv 2602.18914

What is ATO?

ATO (Agent Tool Optimization) is to the agent economy what SEO was to the search economy.

SEO	LLMO	ATO
Target	Search engines	LLM responses	Agent tool selection
Trigger	Human searches	Human asks AI	Agent acts autonomously
Result	A click	A mention	A transaction

LLMO is Stage 1 of ATO — necessary but not sufficient.

Quick Start

Score in browser

**toolrank.dev/score** — paste your tool JSON or enter your Smithery server name.

Score via CLI

bash

npx @toolrank/mcp-server

Score in Python

python

from toolrank_score import score_server, format_report

result = score_server("my-server", tools)
print(format_report(result))

ToolRank Score

0-100 metric across four dimensions:

Dimension	Weight	What it measures
Findability	25%	Can agents discover you?
Clarity	35%	Can agents understand you?
Precision	25%	Is your schema precise?
Efficiency	15%	Are you token-efficient?

Maturity Levels

Level	Score	Meaning
Dominant	85-100	Agents prefer your tool
Preferred	70-84	Agents can use your tool well
Selectable	50-69	Agents might use your tool
Visible	25-49	Agents see you but rarely select
Absent	0-24	Agents can't find you

Before and After

diff

- "name": "get",
- "description": "gets data from the api"
+ "name": "search_repositories",
+ "description": "Searches for GitHub repositories matching a query.
+   Useful for finding open-source projects or checking if a repo exists.
+   Returns name, description, stars, language, and URL.",
+ "inputSchema": {
+   "type": "object",
+   "properties": {
+     "query": { "type": "string", "description": "Search query" },
+     "sort": { "type": "string", "enum": ["stars", "forks", "updated"] }
+   },
+   "required": ["query"]
+ }

Score: 52 → 96. Five minutes of work. 3.6x selection advantage.

Architecture

code

toolrank/
├── packages/
│   ├── scoring/           # Level A engine (Python, zero-cost)
│   │   ├── toolrank_score.py    # 14 checks across 4 dimensions
│   │   ├── level_c_score.py     # Claude AI scoring (Pro)
│   │   └── weights.json         # Auto-calibrated weights
│   ├── scanner/           # Ecosystem scanner
│   │   ├── scanner_v3.py        # Weekly full / daily diff
│   │   ├── calibrate.py         # Weight auto-adjustment
│   │   └── auto_blog.py         # Daily article generation
│   ├── web/               # Astro site (toolrank.dev)
│   ├── mcp-server/        # ToolRank MCP Server
│   └── badge-worker/      # Dynamic badge SVG (CF Workers)
└── .github/workflows/     # Automated pipelines

Ecosystem Rankings

Updated weekly. Full ranking →

Rank	Server	Score
1	microsoft/learn_mcp	96.5
2	docfork/docfork	96.5
3	brave	94.7
4	LinkupPlatform/linkup-mcp-server	93.5
5	smithery-ai/national-weather-service	93.3

Add Badge to Your README

markdown

[![ToolRank](https://toolrank.dev/badge/dominant.svg)](https://toolrank.dev/ranking)

Contributing

ToolRank is open source. The scoring logic is fully transparent and auditable.

Report issues: GitHub Issues
Scoring methodology: packages/scoring/toolrank_score.py
Governance: GOVERNANCE.md · CHANGELOG.md
ATO Framework: toolrank.dev/framework

⭐ Star this repo if you find ToolRank useful — it helps others discover it.

License

MIT

Frequently asked questions

What is toolrank?

toolrank is The first ATO (Agent Tool Optimization) platform. Score and optimize MCP tools so AI agents choose yours.

How do I install toolrank?

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 toolrank open source?

Yes — it is hosted on GitHub at https://github.com/imhiroki/toolrank and has 1 stars.

Related MCP tools

servers

Model Context Protocol Servers

90,051 TypeScript

codegraph

Pre-indexed code knowledge graph, auto syncs on code changes, for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, CoPilot, and Hermes Agent — fewer tokens, fewer tool calls, 100% local

69,582 C

goose

an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM

53,903 Rust

mcpacpai+1

metabase

The easy-to-use open source Business Intelligence and Embedded Analytics tool that lets everyone work with data :bar_chart:

49,091 Clojure

analyticsbusinessintelligencedashboard+15

UI-TARS-desktop

The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra

38,840 TypeScript

agentvlmvision+11

cognee

Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory across sessions with a self-hosted knowledge graph engine.

30,433 Python

agent-memoryagent-skillsai+16

Run your own MCP server? See who uses it and what to fix.

Measure it with TrackMCP