What does Typst MCP Server do that Pandoc alone does not?
Beyond LaTeX-to-Typst conversion, it provides Typst syntax validation, PNG rendering, and direct access to Typst documentation – all integrated into AI workflows via MCP tools.
@johannesbrandenburger
基本情報
Typst MCP Server is an MCP (Model Context Protocol) implementation that helps AI models interact with Typst, a markup-based typesetting system. The server provides tools for converting between LaTeX and Typst, validating Typst syntax, and generating images from Typst code.
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"typst-mcp": {
"command": "python",
"args": [
"server.py"
]
}
}
}ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
Typst MCP Server is an MCP (Model Context Protocol) implementation that helps AI models interact with Typst, a markup-based typesetting system. It provides tools for converting between LaTeX and Typst, validating Typst syntax, and generating PNG images from Typst code.
Clone both this repository and the Typst repository, generate the documentation JSON file, install dependencies with uv sync, then run python server.py or use Docker. Configure the server in your AI platform’s MCP settings (e.g., Claude Desktop, Cursor, VS Code Agent Mode) by specifying the command and arguments.
Beyond LaTeX-to-Typst conversion, it provides Typst syntax validation, PNG rendering, and direct access to Typst documentation – all integrated into AI workflows via MCP tools.
You need Python with `uv`, a clone of the Typst repository (for documentation generation), and optionally Docker. Pandoc must be installed for LaTeX conversion.
The Typst documentation is generated locally from the Typst source code and stored as a JSON file (`main.json`) in the `typst-docs` folder of this repository.
The schema of the Typst documentation is not stable and may change at any time; the server must be updated accordingly. Currently all functionality is implemented as MCP tools because other primitives (resources, prompts) are not yet supported in client IDEs like Cursor and VS Code.
The server uses MCP stdio transport (no authentication required) and exposes all features as tools for AI models to call.

Health by Dr Brad works out a health plan from your own results: which screening is due at your age, and evidence-based suggestions, each with the reason and the published guideline or study behind it. Your record is one
Painless E2E Automation for Mobile and Web

DeepMark helps teachers deliver rapid, consistent marking with meaningful feedback for every student — in a fraction of the time. What once took a week, now takes one free period.
A search tool helps dev to solve the naming things problem.
Hindsight: Agent Memory That Learns
コメント