mcptoon

One Place for All Your Agents' MCP Tools and Skills.

Install 1,000 skills and 1,000 MCP tools locally — and never worry about context flooding.

Tools and skills configured in any single agent just work across Cursor, Claude Code, Codex, and 103+ desktop & terminal agents.

Pioneering on-demand schema retrieval out of the box — no more stuffing context windows, saving 99% context tokens .

Zero dependencies, pure local, offline-first.

If this saves you tokens, please star the repo — it helps other builders discover it.

🌐 Official Website · 🧮 30s Token Tax Calculator · 🇨🇳 中文文档 · 📖 Developer Docs

Skeptical of a synthetic 255-tool sample? Here are 7 real public MCP servers (209 tools) measured on a real machine → — 640,606 → 905 tokens (−99.86%). Or run mcptoon bench on your own catalog.

https://github.com/user-attachments/assets/0272d3a8-c74f-428d-8559-fbf804ce0685

① 💥 Tools, skills & docs eating your context — afraid to add more?

Add as many as you like. Tool schemas 72k → 581 tokens .

Tools −99.2% · Skills −99.99% · Docs 40k→17k per question

② 🧩 Several agents — JSON configs and skills both a chore?

It's a native CLI : any agent that can run a shell command gets every feature — no MCP server, no config. sync covers the GUI clients too ( 103+ ).

No JSON editing, no restart

③ 🔍 Overwhelmed by thousands of MCP tools across the web?

No more hunting for install docs. It first scans and takes over the tools you already have , then finds the rest by plain language — install and call on the spot.

mcptoon discover (scan installed) → search (find online) → call (run directly)

Also: OpenAPI / GraphQL passthrough · OAuth · MCP resources & prompts · saved connections · code sandbox — see All commands .

pip install mcptoon && mcptoon quickstart — 565KB, zero dependencies, zero config.

Stays current on its own: it tells you when a newer version is out, and mcptoon upgrade updates in one command.

🔍 Common Pain Points & Solutions at a Glance (Click to expand)

Your Pain Point / Search Intent Vanilla MCP Behavior mcptoon's Solution Context Window Bloat Adding 10+ tools quickly hits context limits Sends full JSON Schema of every tool every turn (70k+ tokens) Compact Name Index : ~2.2 tokens/tool on idle, schemas fetched on demand (99.2% token saving) Multi-Agent Sync Hell Rewriting mcpServers JSON for Cursor, Claude Code, Windsurf Manual copy-pasting into each client's config file One-Command Takeover : mcptoon sync --takeover automatically discovers and configures 103+ clients Headless CLI Execution Need to call/test MCP tools directly in terminal/scripts Requires spinning up full desktop agents or writing client code Native Standard Library CLI : mcptoon call <server> <tool> with zero dependencies Model Confusion & Hallucination Too many candidate tools dilute model attention Floods model with dozens of complex schemas simultaneously Two-Stage Selection : mcptoon select "<task>" pre-filters to the top 3 tools offline Agent Skills Fragmentation Prompt instructions & skills can't be shared across agents Manually duplicated across folders, drifting out of sync Symlink-based Single-Source Hub : mcptoon skills sync , 39-token pointer, zero duplication

---

🚀 Quick Start (Choose Your Way)

🤖 Modern Way: Just Talk to Your Agent (Zero Memorization)

When installed, mcptoon automatically injects its own official skill (`/mcptoon`) into your agents. You never need to memorize terminal commands — just ask directly in Cursor, Claude Code, Windsurf, or DSH:

# Ask the built-in skill anytime:

"Load /mcptoon skill and check what MCP tools I currently have." "Help me find an MCP tool that converts HTML to PDF and call it." "The tool output was truncated, retrieve the full original copy." "How many tokens did mcptoon save for me today?" "Doctor self-check: are all my MCP servers connected properly?"

*Your agent loads `/mcptoon` and drives everything behind the scenes!*

⚡ Terminal Way: 30-Second Setup

# 1. Install (zero dependencies, 565KB)

pip install mcptoon

2. One command: detect MCP servers & auto-configure agents

(Also self-installs /mcptoon skill into all detected agents)

mcptoon quickstart

3. View compact tool catalog (-99.2% tokens)

mcptoon manifest

4. Route tasks to skills or doc chapters (<5ms)

mcptoon skills resolve "write git commit message" mcptoon docs resolve "how does release gating work"

Is it actually saving you anything? After a fresh install mcptoon can sit in direct mode — installed but not wired into the agent — and save nothing. See Which mode is cheapest?, or just run mcptoon mode.

⚡ Core Capabilities at a Glance

| • ✅ Compact name index (−99.2%): 255 tools → 581 tokens, full schema fetched on demand • ✅ Lossless result compression: reversible, retrieve the original behind any handle • ✅ Session dedup: byte-identical output in one turn becomes a reference • ✅ Skills the same way: one 39-token pointer + one lookup replaces every SKILL.md • ✅ Docs too: docs feeds 1 chapter, not 19 — 40k → 17k tokens per question • ✅ Honest numbers: real tiktoken accounting via mcptoon bench | • ✅ Native CLI: any shell-capable agent runs mcptoon and gets every feature — no MCP server, no config • ✅ GUI clients too: sync writes the JSON for Claude, Cursor, Cline, Windsurf, VS Code, Claude Code — 6 here, 103+ supported • ✅ Verified list: 42 clients run the CLI directly, 61 mount one mcptoon serve entry • ✅ Skills too: one catalog distributed to every agent's folder (skills sync) • ✅ Import what you have: pull servers from Claude / Cursor / Cline / Windsurf • ✅ One-command setup: quickstart = discover + configure + show tools • ✅ Fine control: toggle on/off, policy sets each tool's output tier, --version-gate / --tombstone for skills |

| • ✅ Cross-registry search: plain-language search across 17,000+ MCP servers • ✅ Discover what's already installed: discover scans existing client configs — nothing to reinstall • ✅ Call anything: call headless, or call --auto to auto-find the server • ✅ On-demand schema: inspect only when a tool is actually used • ✅ Pick for a task: select " " returns just the top few • ✅ Export anywhere: manifest --format openai\|openapi\|mcp | • ✅ OpenAPI / Swagger: any REST API as commands (openapi list\|call) • ✅ GraphQL: any endpoint as commands (graphql list\|call) • ✅ OAuth built in: PKCE login, client-credentials, Dynamic Client Registration • ✅ MCP resources & prompts: resources, prompts, complete • ✅ Saved connections: bake — secrets by reference only (env: / file:) • ✅ Code sandbox: exec — only stdout enters context |

| • ✅ Zero dependencies: pure standard library, pyproject.toml clean • ✅ 565KB, installs in 2s • ✅ 100% local & private: no cloud, no teleme