Txtai Assistant

Model Context Protocol (MCP) server implementation for semantic vector search and memory management using TxtAI. This server provides a robust API for storing, retrieving, and managing text-based memories with semantic vector database search capabilities. You can use Claude and Cline AI as well.

python

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "rmtech1-txtai-assistant-mcp": {
      "command": "path/to/txtai-assistant-mcp/scripts/start.sh",
      "args": []
    }
  }
}

You deploy and run the TxtAI Assistant MCP server to store, search, and manage memories with semantic understanding. It exposes a simple programmatic interface that lets you persist memories, retrieve them by meaning, tag and organize content, and monitor health and statistics for reliable memory-powered conversations.

How to use

You will run the MCP server locally and connect your MCP client to it. Use the server to store memories with content, metadata, and tags, then search and retrieve memories using semantic queries or tag filters. You can check health and get statistics to monitor the memory store, and you can perform actions like deleting memories by their content hash when needed.

How to install

Prerequisites you need before installing: Python 3.8 or higher, pip, and virtualenv (recommended). You should also have a terminal or command prompt and git available.

Step 1: Clone the project repository.

git clone https://github.com/yourusername/txtai-assistant-mcp.git

Step 2: Change into the project directory.

cd txtai-assistant-mcp

Step 3: Start the server using the provided startup script.

./scripts/start.sh

Configuration and notes

Configuration is done via environment variables, with a template provided for guidance. The template includes common server settings such as host, port, CORS, logging level, and memory limits.

# Server Configuration
HOST=0.0.0.0
PORT=8000

# CORS Configuration
CORS_ORIGINS=*

# Logging Configuration
LOG_LEVEL=DEBUG

# Memory Configuration
MAX_MEMORIES=0

Usage in clients and tools

You can interact with the server through MCP-compatible tools integrated into your client. The server exposes features to store memories, retrieve memories with semantic search, search by tags, delete memories by content hash, and fetch statistics or health status.

Security and reliability

CORS settings are configurable, and file paths are sanitized to prevent directory traversal. Input validation is performed on all endpoints to ensure robust and predictable behavior.

Troubleshooting

If the server fails to start, verify that you have Python 3.8+ installed, the necessary dependencies, and that the script has execute permissions. Check the logs directory for server.log to see startup details and any errors.

Notes

This MCP server integrates semantic search and memory management using a Python-based backend. It is designed to be used with MCP clients like Claude and Cline to enhance their memory-aware capabilities.

Available tools

store_memory

Store new memory content with metadata and tags so it can be retrieved semantically later.

retrieve_memory

Retrieve memories using semantic search queries to find relevant content even if wording differs.

search_by_tag

Find memories by specific tags to quickly access grouped content.

delete_memory

Remove a memory by its content hash to manage storage and privacy.

get_stats

Obtain database statistics such as memory counts and tag distribution.

check_health

Check the health of the database and the embedding model to ensure reliability.

Built by

VeilStrat

AI signals for GTM teams