Local document knowledge base with CLI and MCP server

Open-source tool for a local knowledge base across 90+ document formats, accessible via CLI and MCP server, no cloud or API keys.

View on GitHub→

Local Document Knowledge Base with CLI and MCP Server

This open-source tool provides a robust local knowledge base solution for developers. It allows you to ingest and query information from over 90 different document formats directly on your machine, eliminating the need for cloud services or external API keys. This makes it ideal for managing sensitive data or for environments with limited internet connectivity. The system is accessible through a command-line interface (CLI) and also exposes an MCP server, facilitating integration with other development workflows.

What it Does

The core functionality is to create and maintain a searchable knowledge base from your local documents. It handles a wide array of file types, parsing their content for indexing. Once indexed, you can query this knowledge base using natural language or specific keywords. The tool abstracts away the complexities of document parsing and semantic search, providing a straightforward interface for developers to leverage their own data.

Key Features

Extensive Document Format Support: Ingests and processes over 90 different document formats.
Local-First Operation: No cloud dependency, all data and processing remain on your local system.
CLI Access: Direct interaction and querying via a command-line interface.
MCP Server: Exposes an MCP server for programmatic access and integration.
Open-Source: Freely available and modifiable under an open-source license.
No API Keys Required: Simplifies setup and usage by removing external service dependencies.

Who it's For

This tool is designed for AI builders, developers, and researchers who need to create and manage private, local knowledge bases. It's particularly useful for:

Developers working with sensitive data: Where cloud storage is not an option.
Teams requiring offline capabilities: For projects in remote locations or with unreliable internet.
Anyone building AI agents or applications: That need to access and reason over a specific set of documents.
Individuals looking for a self-hosted LLM knowledge solution: Without the overhead of complex cloud setups.