"Boost" matching MCP servers:

A local-first document retrieval MCP server that enables AI coding tools like Codex to search private local documents via semantic search and keyword boost, supporting ingestion of PDF, DOCX, TXT, Markdown, and HTML files.

Reduces token costs for AI coding agents by indexing code symbols and returning only relevant functions/classes for a given task.

Enables AI agents to use a hybrid dense+sparse RAG system with tag-based long-term memory (importance, knowledge-type decay, access-frequency boost) via MCP tools for document ingestion, hybrid search, staged reranking, forgetting, and index management.

It lets MCP clients such as Claude Desktop ask natural-language questions about the Model Context Protocol and get grounded answers with reranking, semantic caching, guardrails, PII masking, and hallucination detection.