TeXFlow

A LaTeX document compiler with anMCPinterface. AI agents operate on a structured document model — sections, paragraphs, figures, tables — while TeXFlow handles all LaTeX mechanics: packages, preamble, fonts, and compilation.

# Arch pacman -S texlive-xetex texlive-fontsrecommended texlive-fontsextra # Debian/Ubuntu apt install texlive-xetex texlive-fonts-recommended texlive-fonts-extra # Fedora dnf install texlive-xetex texlive-collection-fontsrecommended texlive-collection-fontsextra

claude mcp add texflow -- uvx texflow-mcp

claude mcp add texflow -- uvx texflow-mcp ~/Documents/TeXFlow

{ "mcpServers": { "texflow": { "command": "uvx", "args": ["texflow-mcp"] } } }

TeXFlow has two entry flows, both converging on the same in-memory document model:
- Scaffold—document(action="create")builds an empty document skeleton. Add content withedit(action="insert").
- Markdown ingest—document(action="ingest", source="paper.md")parses markdown into the model. Refine layout from there.

The model auto-saves to disk as JSON. LaTeX is only ever an output artifact — you never edit.texdirectly.

Every response includes a workflow state hint showing where you are and what to do next.

> document(action="create", title="My Paper", document_class="article") > queue(operations=[ {"tool": "edit", "action": "insert", "block_type": "section", "title": "Introduction", "level": 1}, {"tool": "edit", "action": "insert", "content": "This paper explores...", "section": "Introduction"}, {"tool": "edit", "action": "insert", "block_type": "section", "title": "Methods", "level": 1}, {"tool": "layout", "font": "palatino", "columns": 2} ]) > render(action="compile")

git clone https://github.com/aaronsb/texflow-mcp cd texflow-mcp uv sync uv run pytest tests/ -v # 222 tests uv run texflow # Start MCP server

Collaborative word processor you can use with your agent

Universal document generation and conversion MCP. Generate PDF/DOCX/XLSX from templates+JSON (invoices, contracts, reports), batch generation, 100+ format conversions.

Carryo is a remote MCP server for sharing Claude or ChatGPT-created HTML artifacts as live links.

An MCP server that enables AI assistants to interact with Confluence content through a standardized interface.

Generates professional PDF cover letters using LaTeX. Requires Docker for local execution.

Publish Markdown or HTML to a clean, shareable link straight from your AI assistant — OAuth, no API keys.

A server for reading and converting documents between PDF, DOCX, and Markdown formats using marker-pdf and pandoc.

Search, cite, and read PDFs from your EndNote reference library directly in Claude Desktop

An MCP server for interacting with the esa.io knowledge-sharing platform.