Continuo Memory

Continuo is a persistent memory system that provides semantic search and storage capabilities for development workflows. By separating reasoning (LLM) from long-term memory (Vector DB + hierarchical compression), the system maintains knowledge indefinitely, circumventing context window limitations.

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": {
    "gtokai-continuo-memory-mcp": {
      "command": "python",
      "args": [
        "src/mcp/memory/mcp_memory_server.py",
        "--provider",
        "local",
        "--db-path",
        "./memory_db"
      ]
    }
  }
}

Continuo Memory MCP Server provides persistent, local memory with semantic search and hierarchical compression that stays available across sessions. It integrates with IDEs through the Model Context Protocol, enabling you to store, retrieve, and reason about development knowledge efficiently while keeping costs low and data private.

How to use

You run the memory server locally and connect your IDE to it so you can search, store, and summarize development knowledge within your workflow.

Start the memory server to enable persistent memory and semantic search.

How to install

Follow these steps to set up the memory MCP server locally and connect it to your IDE.

Prerequisites you need installed on your system include Python 3.9+ and a working Python environment.

Clone the project repository and navigate into it.
Install dependencies for the memory MCP server and the environment.

Additional sections

Configuration notes, security considerations, examples, and troubleshooting help help you maintain a stable memory setup across projects.

Security: Keep local memory data private by running entirely on your workstation or a trusted internal network. Never expose the memory store to untrusted networks unless you intend to use a controlled, secure deployment.

Examples: Use basic store and search operations to verify that memory chunks are correctly stored, summarized, and retrievable across N0, N1, and N2 levels.

Tools and endpoints

The MCP tools exposed enable you to search, store, and inspect memory, along with memory statistics.

Available tools

search_memory

Semantic search in persistent memory to find relevant documents and compute similarity scores

store_memory

Store content in persistent memory with optional metadata and hierarchical level tagging (N0, N1, N2)

get_memory_stats

Retrieve memory usage statistics such as total documents and level distribution

Built by

VeilStrat

AI signals for GTM teams