[integrations]

Integrations

Connect shodh-memory to your favorite AI coding tools. One config, persistent memory across every session.

claude-code.config
[claude-code]

Claude Code

The fastest way to get started. One CLI command registers shodh-memory as an MCP server with all 45 tools. Pair with hooks for fully automatic memory capture across sessions.

# setup
  1. 01.Run the command below in your terminal
  2. 02.All 45 MCP tools are available immediately
  3. 03.Add hooks for automatic memory (see note below)
terminal
claude mcp add shodh-memory -- npx -y shodh-memory
Works with Claude Code hooks for automatic memory capture. Add hook configuration to ~/.claude/settings.json to record context from every session, tool use, and prompt without manual tool calls.
cursor.config
[cursor]

Cursor

Add persistent memory to Cursor's AI assistant. Memories persist across chat sessions, composer threads, and projects.

# setup
  1. 01.Open Cursor Settings (Cmd/Ctrl + ,)
  2. 02.Navigate to MCP section
  3. 03.Click "Add new MCP server"
  4. 04.Paste the JSON config below
~/.cursor/mcp.json
{
  "mcpServers": {
    "shodh-memory": {
      "command": "npx",
      "args": ["-y", "shodh-memory"]
    }
  }
}
windsurf.config
[windsurf]

Windsurf

Give Windsurf's Cascade AI long-term memory. Remembers decisions, patterns, and context across coding sessions.

# setup
  1. 01.Open Windsurf Settings
  2. 02.Navigate to MCP Servers
  3. 03.Click "Add Server"
  4. 04.Paste the JSON config below
mcp_config.json
{
  "mcpServers": {
    "shodh-memory": {
      "command": "npx",
      "args": ["-y", "shodh-memory"]
    }
  }
}
claude-desktop.config
[claude-desktop]

Claude Desktop

Add persistent memory to the Claude Desktop app. Memories carry across conversations so Claude remembers your preferences and project context.

# setup
  1. 01.Locate your Claude Desktop config file (see paths below)
  2. 02.Add the shodh-memory server config
  3. 03.Restart Claude Desktop
claude_desktop_config.json
{
  "mcpServers": {
    "shodh-memory": {
      "command": "npx",
      "args": ["-y", "shodh-memory"]
    }
  }
}
Config file locations: macOS ~/Library/Application Support/Claude/claude_desktop_config.json | Windows %APPDATA%\Claude\claude_desktop_config.json | Linux ~/.config/Claude/claude_desktop_config.json
continue.config
[continue]

VS Code + Continue

Use shodh-memory with the Continue extension in VS Code. Gives your open-source AI assistant persistent memory across sessions.

# setup
  1. 01.Install the Continue extension in VS Code
  2. 02.Open Continue settings (continue/config.json)
  3. 03.Add shodh-memory to the MCP servers section
  4. 04.Reload VS Code
~/.continue/config.json
{
  "mcpServers": [
    {
      "name": "shodh-memory",
      "command": "npx",
      "args": ["-y", "shodh-memory"]
    }
  ]
}
docker.config
[docker]

Docker

Run shodh-memory as a standalone server for self-hosted deployments, CI/CD pipelines, team-shared memory, or multi-agent architectures.

# setup
  1. 01.Pull and run the Docker image
  2. 02.Memory data persists in the shodh-data volume
  3. 03.Access the REST API at http://localhost:3030
  4. 04.Point any MCP client at the running server
terminal
docker run -d -p 3030:3030 \
  -v shodh-data:/data \
  ghcr.io/varun29ankus/shodh-memory:latest
For CI/CD: mount a persistent volume so memory survives container restarts. Set SHODH_HOST=0.0.0.0 to accept connections from other containers on the same network.
compatibility.md

Works with any MCP client

shodh-memory implements the Model Context Protocol (MCP) standard. Any MCP-compatible client can connect to shodh-memory out of the box — no custom adapters, no vendor lock-in.

# The universal pattern — works with any MCP client:
#
#   command:  npx
#   args:     ["-y", "shodh-memory"]
#
# That's it. 45 tools. Zero configuration.
# Memory that learns with use, runs locally, never phones home.

Pick your tool, paste the config, and start building with memory.

Get StartedView Source