Cursor10x

Name: Cursor10x
Author: aurda012

@aurda012

About Cursor10x

The Cursor10x MCP is a persistent multi-dimensional memory system for Cursor that enhances AI assistants with conversation context, project history, and code relationships across sessions.

#memory #sqlite #mcp #vector #cursor #vector-database #turso #agentic #agentic-workflow #agentic-ai #cursorai #cursor-ai #mcp-server #agent-memory #agent-db

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "cursor10x-mcp": {
      "command": "npx",
      "args": [
        "cursor10x-mcp"
      ],
      "enabled": true,
      "env": {
        "TURSO_DATABASE_URL": "your-turso-database-url",
        "TURSO_AUTH_TOKEN": "your-turso-auth-token"
      }
    }
  }
}

Tools

Generates a banner containing memory system statistics and status

Checks the health of the memory system and its database

Initializes a conversation by storing the user message, generating a banner, and retrieving context in one operation

Ends a conversation by storing the assistant message, recording a milestone, and logging an episode in one operation

Stores a user message in the short-term memory

Stores an assistant message in the short-term memory

Tracks an active file being accessed by the user

Retrieves recent messages from the short-term memory

Retrieves active files from the short-term memory

Stores a project milestone in the long-term memory

Stores a project decision in the long-term memory

Stores a project requirement in the long-term memory

Records an episode (action) in the episodic memory

Retrieves recent episodes from the episodic memory

Retrieves comprehensive context from all memory systems

Retrieves statistics about the memory system

Unified tool for managing vector embeddings with operations for store, search, update, and delete

Run diagnostics on the vector storage system to identify issues

Overview

What is Cursor10x?

Cursor10x (now evolved into DevContext) is a Model Context Protocol (MCP) server that creates a persistent memory layer for AI assistants, specifically Claude. It enables assistants to retain and recall conversation history, active files, project milestones, decisions, technical requirements, code snippets, and semantically related content across sessions.

How to use Cursor10x?

Install Node.js 18+ and create a Turso database account. Configure .cursor/mcp.json in your project with the Turso database URL and auth token. Then run the server via npx cursor10x-mcp. The AI assistant invokes the provided tools automatically.

Key features of Cursor10x

Persistent context across multiple sessions
Importance-based information prioritization
Multi-dimensional memory (STM, LTM, Episodic, Semantic)
Vector embeddings for semantic similarity search
Automatic code indexing and structure detection
Health monitoring and built-in diagnostics

Use cases of Cursor10x

AI assistant recalling past decisions and milestones during development
Searching for related code across files by semantic meaning
Keeping conversation context across multiple chat sessions
Tracking project progress through episodic memory of actions

FAQ from Cursor10x

What is the relationship between Cursor10x and DevContext?

Cursor10x has evolved into DevContext, a more powerful dedicated context system for developers. The repository and documentation refer to the original name.

What dependencies does Cursor10x require?

Node.js 18 or higher and a Turso database account. The free Turso plan is sufficient for typical projects.

What tools does Cursor10x provide?

It provides tools for initializing and ending conversations, checking health, retrieving memory stats, and managing short-term, long-term, episodic, and semantic memory. Key tools include mcp_cursor10x_initConversation, mcp_cursor10x_endConversation, mcp_cursor10x_checkHealth, mcp_cursor10x_getMemoryStats, and mcp_cursor10x_getComprehensiveContext.

How is data stored in Cursor10x?

All memory data is persisted in a Turso database with automatic schema creation. Vector embeddings are stored using Turso's F32_BLOB and vector functions for efficient similarity search.