haiku.rag MCP Server
ggozad/haiku.rag
[](https://agentrank-ai.com/tool/ggozad--haiku.rag/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge) Agentic RAG for local and self-hosted document search: hybrid retrieval, reranking and multimodal RAG on embedded LanceDB, with Docling parsing and an MCP server
claude mcp add agentrank -- npx -y agentrank-mcp-server Overview
ggozad/haiku.rag is a Python MCP server licensed under MIT. Agentic RAG for local and self-hosted document search: hybrid retrieval, reranking and multimodal RAG on embedded LanceDB, with Docling parsing and an MCP server Topics: mcp-server, rag, lancedb, pydantic-ai, docling, ag-ui, agentic-rag, code-execution, code-mode, cross-encoder, document-ai, document-ingestion, document-parsing, hallucination-prevention, multimodal-rag, offline-llm, rag-evaluation, reciprocal-rank-fusion, reranking.
Ranked #179 out of 24842 indexed tools.
In the top 1% of all indexed tools.
Has 12 contributors.
Actively maintained with commits in the last week.
Ecosystem
Score Breakdown
624 stars → growing community
Last commit today → actively maintained
166/170 issues closed → responsive maintainer
12 contributors → active community
No dependents → no downstream usage
1,043 weekly installs → gaining traction
Weights: Freshness 20% · Issue Health 20% · Dependents 22% · Stars 10% · Contributors 8% · How we score →
Badge all embed codes →
[](https://agentrank-ai.com/tool/ggozad--haiku.rag/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge) <a href="https://agentrank-ai.com/tool/ggozad--haiku.rag/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge"><img src="https://agentrank-ai.com/api/v1/badge/ggozad--haiku.rag" alt="AgentRank"></a> Matched Queries
From the README
# haiku.rag Agentic RAG that answers questions about your own documents with citations to page numbers and section headings. It runs on an embedded LanceDB database with open models through Ollama by default, so no server or API key is needed. Any provider Pydantic AI supports works in their place, and the same database can live on S3, GCS, Azure or LanceDB Cloud. Built on [LanceDB](https://lancedb.com/), [Pydantic AI](https://ai.pydantic.dev/) and [Docling](https://docling-project.github.io/docling/). Documentation: [ggozad.github.io/haiku.rag](https://ggozad.github.io/haiku.rag/). ## Features - **Ingest** PDFs, office documents, HTML, Markdown and images with Docling, in-process or on docling-serve. The stored DoclingDocument keeps headings, tables, pictures and page provenance. - **Search** with hybrid vector and full-text retrieval, optional reranking (cross-encoders, Jina, Cohere, Zero Entropy, vLLM, OpenRouter), section-aware context expansion, and image search with a multimoRead full README on GitHub →
Get the weekly AgentRank digest
Top movers, new tools, ecosystem insights — straight to your inbox.