Skip to main content
Glama
AI Knowledge Base MCP Server

AI Knowledge Base MCP Server

by Alpha-W0lf

AI Knowledge Base

Production-grade local hybrid RAG that keeps coding agents current: cited passages from your own documents, served through a read-only MCP (Model Context Protocol) server. LanceDB stores the index and Ollama makes the embeddings, so no documents or queries leave your machine. The first reranker load downloads its model weights from Hugging Face.

The public repo runs on committed synthetic documents.

CI Eval

Proof

A pre-registered evaluation: 48 synthetic, AI-written documents in 12 look-alike clusters, 96 graded queries, and a held-out set written blind by a separate agent that saw only the documents. The pre-registration fixes the data by hash, and the results file records the same hashes (pre-registration, results).

Held-out, 48 queries

MRR@10

nDCG@10

hit@1

Hybrid search + cross-encoder (default)

0.948

0.962

0.896

Hybrid search (RRF fusion, no reranker)

0.925

0.952

0.875

Keyword search alone

0.922

0.939

0.896

Vector search alone

0.893

0.922

0.812

Random order

0.061

0.096

0.021

The right document comes back first on 90% of held-out queries, against 2% by chance, and the full pipeline scores higher than either search leg alone on MRR@10 and nDCG@10 (against keyword-only, the margin is within the 95% confidence interval). Measured 2026-10-18 with nomic-embed-text and ms-marco-MiniLM-L-6-v2; the results file records the model revisions and dataset hashes. This repo started as a single-commit snapshot of the development history, so the commit named in the results file is from that pre-release history. On every ranking change, CI reruns the eval on its offline embedder and fails the build if a hybrid-search metric drops more than 0.02 below the committed baseline.

Related MCP server: Keepygaga RAG

The problem

AI techniques move weekly, and coding agents that only know their training data fall behind. Teams need a local, citable knowledge path: retrieve what matters, cite the source, and give agents tools they can call safely, without sending private documents to a third party.

How it works

flowchart LR
  F[Documents] --> I[Chunk + embed]
  I --> D[(LanceDB<br/>vectors + full-text index)]
  Q[Query] --> V[Vector search]
  Q --> K[Keyword search]
  D --> V
  D --> K
  V --> R[RRF fusion]
  K --> R
  R --> C[Cross-encoder rerank]
  C --> O[Cited results<br/>CLI or MCP]

Meaning-based (vector) search and keyword (full-text) search run side by side. Reciprocal Rank Fusion (RRF) merges the two lists, and a cross-encoder reranker re-scores the top results. CLI, MCP and the eval all call the same retrieve() function, and every result carries its citation (source_id, source_url), its rank in each leg and the stage that produced the final order.

Key engineering decisions

  1. Hybrid fusion before the cross-encoder. Search stays fast and useful when the reranker is slow, missing or broken: a failed reranker falls back to the fused ranking and says so (ranking_stage: fusion_degraded). Hybrid search stops with a clear error if the keyword index is missing, and an index built with a different embedding model is refused until you rebuild it.

  2. Public fixtures, private corpus. The repo ships synthetic documents with a license manifest. Your own channels and transcripts stay in git-ignored files, and every citation uses a stable source_id, never a local file path.

  3. MCP read-only by default. The public profile exposes 4 tools, each advertising readOnlyHint. Search never writes: ingest builds the index, and a test proves that searching leaves the database files untouched. Tools that change data exist only when AI_KB_MCP_PRIVATE=1 is set.

Try it

You need Python 3.11+, uv and Ollama.

uv sync --extra ce                       # dependencies + cross-encoder
ollama pull nomic-embed-text             # local embedding model
uv run python -m src.ingest --fixtures   # index the smoke documents
uv run python -m src.search "reciprocal rank fusion RRF" --hybrid --ce
uv run python -m src.eval                # the 48-document evaluation

Set AI_KB_EMBEDDINGS=stub to run every command offline on the CI embedder (add --no-ce to search and eval or set AI_KB_CE=0, or download the reranker once first). To connect an agent, copy mcp-config.example.json into your MCP client and set cwd to your clone. Audit the server with mcp-audit: mcp-audit run --server "uv run python -m src.mcp_server".

Stack

Python 3.11+ · LanceDB 0.27 · Ollama nomic-embed-text (768-dim) · sentence-transformers cross-encoder ms-marco-MiniLM-L-6-v2 · MCP Python SDK 2.3 · uv

Deeper docs

Built by Thomas (Tom) Chacko, Senior AI Engineer (Ex-Meta). Building agent knowledge systems? Reach me on LinkedIn.

Available Tools

4 tools
discoverA
Read-only

Discover AI content without a specific query (honest browse/digest/heuristics).

Modes: random, digest, concepts, channels — not a ranking product. Calls shared implementation in src.discover.

Inputs: days, mode, limit. Returns: result

get_contextC
Read-only

Get comprehensive context via shared retrieval (+ optional recent digest).

Inputs: query (required), include_recent. Returns: result

get_statusA
Read-only

Get current status of the AI knowledge base.

Does not require embedding / Ollama for index stats (Ollama health is reported).

Inputs: no named parameters. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3.2/5.0

    Scored across 4 tools

    Disambiguation3/5

    get_context, search, and discover all relate to retrieving information, which creates some overlap; however, search is query-based, discover is browse-based, and get_context aims for comprehensive retrieval, so descriptions mostly disambiguate. get_status is clearly separate. The boundary between get_context and search remains slightly fuzzy.

    Naming Consistency3/5

    Two tools use the get_ prefix while two use bare verbs, creating a mixed convention. The names are still readable and each clearly indicates its action, but the inconsistent style is noticeable.

    Tool Count4/5

    Four tools is a reasonable size for a focused knowledge-base retrieval server, though it feels on the lean side. Each tool has a distinct role and the count is appropriate for the apparent scope.

    Completeness2/5

    The server only provides retrieval and status operations; there are no tools for adding, updating, or deleting knowledge-base content. For a server named 'AI Knowledge Base', this is a significant gap that limits agents to read-only workflows.

    Maintenance

    ActivityMaintained
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      Not graded
      quality
      B
      maintenance
      Provides read-only access to a personal RAG knowledge base, enabling hybrid search, evidence-grounded retrieval with citations, and knowledge gap tracking for LLM agents.
      -
    • A
      license
      A
      quality
      C
      maintenance
      Enables local-first hybrid knowledge retrieval from authorized Markdown and plain-text files, combining full-text and vector search with reranking and traceable source references via a single search tool.
      1
      MIT
    • F
      license
      Not graded
      quality
      A
      maintenance
      Enables AI agents and MCP hosts to query locally indexed notes, docs, chat logs, and other files via hybrid vector/BM25 retrieval with section-level citations. Exposes tools, resources, and prompts for searching and asking over the indexed content.
      -
    • F
      license
      Not graded
      quality
      A
      maintenance
      Enables AI agents and MCP hosts to query a local index of scattered notes and documents through hybrid vector and BM25 retrieval, returning section-level citations and exposing tools, resources, and prompts.
      -