Deadzone

laradji/deadzone

A privacy-focused semantic search engine for library documentation that runs entirely on your machine. Uses nomic-embed-text-v1.5 embeddings stored in a local Turso database with cosine similarity search. Exposes two MCP tools: search_libraries() to find relevant packages by name, and search_docs() to query documentation snippets with natural language and get token-budgeted responses. Ships as a single Go binary with embedded ONNX runtime, no Python or Node dependencies. The container image bakes the database and model weights so it runs offline from first launch. Reach for this when you want grep-beating documentation search without sending queries to external APIs or managing API keys.

GitHub stars: 2
tools: 2
transport: STDIO
last updated: May 7, 2026

View on GitHub

Tools

2 tools

Public tool metadata for what this MCP can expose to an agent.

search_librariesFree text search matched against the libs table, returning library matches with lib_id, version, doc_count and match_score.

Free text search matched against the libs table, returning library matches with lib_id, version, doc_count and match_score.

No parameter schema in public metadata yet.

search_docsSemantic search over indexed documentation using natural language queries, returning token-budgeted snippets.

Semantic search over indexed documentation using natural language queries, returning token-budgeted snippets.

No parameter schema in public metadata yet.

Readme

         _                _
      __| | ___  __ _  __| |_______  _ __   ___
     / _` |/ _ \/ _` |/ _` |_  / _ \| '_ \ / _ \
    | (_| |  __/ (_| | (_| |/ / (_) | | | |  __/
     \__,_|\___|\__,_|\__,_/___\___/|_| |_|\___|

    > semantic doc search. local file. no cloud. no key.
    > you ask in english. it answers in snippets.

Status. Vector search wired end-to-end. MCP over stdio. One binary, Linux + macOS, zero telemetry. See releases for the latest tag and the roadmap for in-flight work. The scraper is still the messy half — #64 is honest about it.

The pitch, in one paragraph

Your AI client says "how do I register a tool?". The doc says AddTool. A grep-based index shrugs; a vector index doesn't. Deadzone is the vector index — nomic-embed-text-v1.5 over Turso's native cosine distance, wrapped in a Go binary that speaks MCP over stdio and keeps every byte on your laptop. It is, roughly, Context7 with the internet turned off.

Rules of the deadzone

View the full README on GitHub

Details

Repository: github.com/laradji/deadzone
Website: github.com/laradji/deadzone
Category: Search & Web Crawling
Package: ghcr.io/laradji/deadzone:v0.7.4
Transport: STDIO
Registry: MCP registry, active

Related Search & Web Crawling MCP servers

View all

Parseflowlibres-coder/parseflowPDF parsing server with text extraction, metadata, search, images, and TOC via MCP2stars

Mcp Domain To Linkedin Url Resolvermambalabsdev/mcp-domain-to-linkedin-url-resolverResolves a company domain or name to its LinkedIn URL from search results, with confidence.2stars

AI Context Flowweb3-plurality/plurality-mcp-serverUniversal memory for AI agents and tools. Save, organize and search context anywhere.2stars

ArxSourcenicolaiprodromov/arxsourceSearch, read and analyze arXiv papers, and find their code: LaTeX, citations, alerts.2stars

IndieStackpattyboi101/indiestackSearch 3,100+ developer tools via MCP. Install commands, compatibility data, health scores.2stars

BAILII UK Case Lawpaulieb89/bailii-mcpSearch UK case law on BAILII — court judgments with section extraction. Runs locally.2stars