MCP server that reduces LLM context by removing code comments and converting data formats to TOON
Connect Toon to Claude, Cursor or any other MCP client and it stops being a tab you switch to. MCP server that reduces LLM context by removing code comments and converting data formats to TOON. The toon mcp server is what makes that connection.
A specialized Model Context Protocol (MCP) server that optimizes token usage by converting data to TOON (Token-Oriented Object Notation) and stripping non-essential context from code files.
The toon-parse-mcp MCP server helps AI agents (like Cursor, Claude Desktop, etc.) operate more efficiently by: 1. Optimizing Code Context: Stripping comments and redundant spacing from code files while preserving functional structure and docstrings. 2. Data Format Conversion: Converting JSON, XML, YAML, and CSV inputs into the compact TOON format to save tokens. 3. Mandatory Efficiency Protocol: A built-in resource that instructs LLMs to prioritize token-saving tools.
The toolset is worth reading before you wire it up, because it tells you what the integration is really for:
Tools — The Tools tool exposed by this serverResources — The Resources tool exposed by this serverCursor — 1. Open Cursor Settings -> MCP. 2. Click "+ Add New MCP Server". 3. Name: toon-parse-mcp 4. Type: command 5. Command: python3 -mWindsurf — 1. Click the hammer icon in the Cascade toolbar and select "Configure". 2. Alternatively, edit ~/.codeium/windsurf/mcp_config.json directly. 3. AddAntigravity — 1. Open the MCP store via the "..." menu at the top right of the agent panel. 2. Select "Manage MCP Servers" -> "View raw config". 3. AlternativelyInstallation goes through your MCP client rather than a global install: point it at toon-parse-mcp on PyPI and it is fetched when the client starts. The copy-paste blocks for Claude Desktop, Claude Code and Cursor are further down this page.
mcp >= 1.25.0 - toon-parse >= 2.4.3This sits in the AI and media services group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Toon's toolset — Tools, Resources, Cursor and 2 more — is a fair guide to whether it matches your workflow. It is maintained by ankitpal181; worth a glance at recent repository activity before you build anything load-bearing on it.
This entry was verified against Toon's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.
| Tool | What it does |
|---|---|
| Tools | The Tools tool exposed by this server. |
| Resources | The Resources tool exposed by this server. |
| Cursor | 1. Open Cursor Settings -> MCP. 2. Click "+ Add New MCP Server". 3. Name: toon-parse-mcp 4. Type: command 5. Command: python3 -m toon_parse_mcp.server (Ensure your environment is active or use absolute path to python) |
| Windsurf | 1. Click the hammer icon in the Cascade toolbar and select "Configure". 2. Alternatively, edit ~/.codeium/windsurf/mcp_config.json directly. 3. Add the following to the mcpServers object: |
| Antigravity | 1. Open the MCP store via the "..." menu at the top right of the agent panel. 2. Select "Manage MCP Servers" -> "View raw config". 3. Alternatively, edit ~/.gemini/antigravity/mcp_config.json directly. 4. Add the followi |
{
"mcpServers": {
"toon-parse": {
"command": "uvx",
"args": ["toon-parse-mcp"]
}
}
}Add to claude_desktop_config.json, then restart Claude Desktop.
mcp >= 1.25.0 - toon-parse >= 2.4.3Build a programmable telecommunications stack for connecting telephony services with the Internet via a cloud-based utility.
Search built for AI, not humans — semantic web search that returns model-ready content, plus code context.
Answers, not links — delegate questions to Perplexity's search-grounded models and get cited responses back.
Give your assistant a voice — text-to-speech, voice cloning and audio tools from the ElevenLabs API.
Give your assistant a real code sandbox — isolated cloud VMs for actually running the code it writes.
The ML hub in your context window — search models, datasets, papers and run Spaces from the official server.