Ingest PDF folders into a local vector database for semantic search and expose them as tools to MCP clients.
Sponsored
DocMCP allows users to transform local PDF document collections into powerful semantic search tools. It ingests PDF folders, generates embeddings using configurable local ONNX models, and stores them in a local ChromaDB vector database. Featuring a user-friendly web UI, DocMCP simplifies the process of ingesting, searching, and managing document collections. It seamlessly integrates with Kiro IDE (or any MCP client) by exposing a suite of tools for semantic search, listing sources, and retrieving full document texts, enabling AI-assisted workflows directly within development environments.
主な機能
010 GitHub stars
02Supports fully local, configurable ONNX embedding models for privacy and offline use.
03Local PDF ingestion into a ChromaDB vector database.
04Exposes semantic search and PDF context tools to MCP clients like Kiro IDE.
05CLI tools for advanced PDF ingestion and collection management.
06Web UI for managing PDF ingestion, semantic search, and collection administration.
ユースケース
01Perform semantic search on personal or project-specific PDF documentation collections.
02Integrate PDF-based knowledge bases directly into AI development environments like Kiro IDE.
03Manage and query custom documentation repositories for AI-assisted coding or research workflows.
