doc2mermaid

Convert documents to visual knowledge maps — powered by LLM + Mermaid.

Document Text → LLM Structure Extraction → Graph JSON → Mermaid → SVG/PNG

Install

pip install doc2mermaid

# Mermaid CLI for SVG/PNG rendering
npm install -g @mermaid-js/mermaid-cli

Quick Start

Python API

from doc2mermaid import doc_to_map

svg_path = doc_to_map(
    open("article.md").read(),
    output="map.svg",
    llm_base_url="https://api.openai.com/v1",
    llm_api_key="sk-...",
    llm_model="gpt-4o-mini",
)

CLI

export DOC2MERMAID_BASE_URL=https://api.openai.com/v1
export DOC2MERMAID_API_KEY=sk-...
export DOC2MERMAID_MODEL=gpt-4o-mini

doc2mermaid article.md -o map.svg
doc2mermaid article.md -o map.png --theme dark
cat article.txt | doc2mermaid - -o map.svg

How It Works

1. Extract: LLM reads your document and extracts a graph structure (nodes + edges)
2. Render: The graph is converted to Mermaid diagram syntax with auto-styling
3. Output: Mermaid CLI renders the final SVG or PNG

Node Types & Colors

Configuration

LLM settings via environment variables or function arguments:

Works with any OpenAI-compatible API (OpenAI, Claude, DeepSeek, local models, etc.).

License

MIT