---

🧠 FHEVM Docs to LLM

This repository provides a Python utility that downloads and compiles the Zama FHEVM documentation into a single Markdown file (zama-llm.txt) — perfect for importing into Cursor, or any other LLM-based development environment that accepts Markdown as context.

📘 Source Repository

The documentation is pulled directly from the official Zama FHEVM project:

🔗 https://github.com/zama-ai/fhevm/tree/main/docs

fhevm (Fully Homomorphic Encryption for EVM) enables private smart contract computation using Zama’s homomorphic encryption technology.
This script helps you integrate that documentation locally for enhanced LLM-assisted coding.

🧩 Why This Exists

When using Cursor (or any AI IDE), it’s useful to provide the LLM with the project’s complete documentation.
Instead of pasting multiple Markdown files manually, this tool:

• ✅ Downloads the entire FHEVM docs/ folder automatically
• ✅ Merges all .md, .mdx, and .txt files into one large, structured Markdown file
• ✅ Adds a Table of Contents with anchor links
• ✅ Outputs a single file — zama-llm.txt — that you can drag & drop into Cursor (or any AI context window)

🚀 Quick Start

1️⃣ Clone this repo

git clone https://github.com/romispectrum/zama-llm-docs.git
cd zama-llm-docs

2️⃣ Run the builder

By default, it downloads from Zama’s fhevm repo and produces zama-llm.txt.

python3 main.py

Output example:

Downloading 'docs/' from zama-ai/fhevm@main ...
Downloaded docs/ to docs
Wrote /path/to/zama-llm.txt with 79 files combined.

3️⃣ Import into Cursor

• Open Cursor
• Go to your project
• Drag and drop zama-llm.txt into your “Knowledge” or “Docs” section
• Cursor will now have full access to the combined FHEVM documentation

⚙️ Command-line Options

Flag	Description	Default
--root, -r	Root folder to scan/download into	./docs
--out, -o	Output file path	./zama-llm.txt
--github	GitHub tree URL to fetch docs from	https://github.com/zama-ai/fhevm/tree/main/docs
--ext	Include extra file extensions (e.g. --ext .rst)	none
--no-default-excludes	Include hidden/system folders (not recommended)	disabled

🧠 Ideal for

• Cursor AI users wanting to give their model the full context of FHEVM docs
• Developers preparing a single Markdown file for LLM fine-tuning or vector embedding
• Offline reference — read all FHEVM docs in one searchable text file

🧑‍💻 Contributing

Contributions are welcome — PRs, bug reports, and improvements appreciated! If you’d like to extend support for other repos or file formats, open an issue.

🪪 License

Released under the MIT License.

❤️ Author

Created by Romi.