Iruziky/docs-forai-data

Docs-ForAI Data 🚀

This dataset contains the pre-computed LanceDB vector indices for the Docs-ForAI project.

It provides high-quality RAG (Retrieval-Augmented Generation) context, allowing AI Agents to access verified information from the most popular AI framework documentations without hallucinations.

📂 Included Documentations

This index currently covers:

1. LangGraph
2. CrewAI
3. PydanticAI
4. AutoGen
5. LlamaIndex
6. OpenAI Swarm
7. Phidata
8. Haystack
9. PraisonAI
10. LiteLLM

🛠 Project Integration

This data is specifically formatted for use with the Docs-ForAI MCP Server.

To use these indices, clone the main repository and run the setup command:

# Clone the project
git clone [https://github.com/iruziky/Docs-ForAI.git](https://github.com/iruziky/Docs-ForAI.git)
cd Docs-ForAI

# Download this dataset automatically
uv run setup

🧠 Technical Specs

• Vector Store: LanceDB
• Embedding Model: sentence-transformers/all-MiniLM-L6-v2 (Running locally on CPU)
• Content: This dataset contains only the computed indices (lancedb_data/). Raw text files (docs_input/) are not included to keep the download efficient.

---

⚖️ Legal Disclaimer

Docs-ForAI is a research and productivity tool. We do not own the documentation content. All rights belong to the respective owners. Our goal is to enhance the developer experience by providing better context to AI models. If you are a copyright holder and want your documentation removed, please open an issue in the Main GitHub Repository.