A Model Context Protocol (MCP) service that extracts text content from files using Apache Tika
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"mcp-file-reader","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# MCP File Reader A Model Context Protocol (MCP) service that extracts text content from files using Apache Tika. This service provides a single tool that can read various file formats (PDF, Word, Excel, PowerPoint, images with OCR, and more) and return their text content. ## Features - **File Content Extraction**: Reads files and extracts text using Apache Tika - **Multiple Format Support**: Supports PDF, DOCX, XLSX, PPTX, images, and many other formats - **Automatic Tika Management**: Automatically starts and manages Tika server when needed - **Simple Deployment**: Easy installation and setup using uv - **Directory Access Control**: Configurable allowed directories for secure file access - **Path Traversal Protection**: Prevents access outside allowed directories via path traversal attacks - **Environment Configuration**: Configurable Tika server endpoint and allowed directories - **Error Handling**: Comprehensive error handling for missing files, network issues, etc. ## Installation and usage Since this package is designed to be used as an MCP service, it is typically installed by inserting an entry into the configuration of your MCP client. The examples below are for the `claude_desktop_confoig.json` settings for the Claude Desktop app, but the general content of the settings should be the same for other applications. ### Using uvx (Recommended) Insert the `"file-reader"` stanza below into your `mcpServers` configuration: ```json { "mcpServers": { "file-reader": { "command": "uvx", "args": [ "mcp-file-reader", "/Users/your_name/Desktop", "/Users/your_name/Downloads", "/Users/your_name/other_accessible_directory" ] } } } ``` ### Running from a local development copy Check out the latest source using: ```bash cd /Users/your_name/source_path git clone https://github.com/nickovs/mcp_file_reader.git ``` Then insert the `"file-reader"` stanza below into your `mcpServers` configuration: ```json …
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.