Install MemoV MCP Fast: A Practical Setup Guide
This guide walks you through installing MemoV and connecting it to your AI coding tools using the Model Context Protocol (MCP). The entire setup takes less than 5 minutes.
Prerequisites
Before installing MemoV, you need two things:
1. Install uv
is a fast Python package manager that MemoV uses for installation:codeuv
macOS / Linux:
bashcurl -LsSf https://astral.sh/uv/install.sh | sh
Windows:
powershellpowershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
2. Install Git
MemoV uses Git's storage format. Install Git if you don't have it:
- macOS: code
brew install git - Ubuntu/Debian: code
sudo apt install git - Windows: Download from git-scm.com
Connect to Your AI Coding Tool
Choose your tool and follow the instructions:
Claude Code
Claude Code is Anthropic's official CLI. Add MemoV from your project directory:
bashcd your-project claude mcp add mem-mcp --scope project -- uvx --from git+https://github.com/memovai/memov.git mem-mcp-launcher stdio $(pwd)
The
flag limits MemoV to this project only.code--scope project
Codex
For OpenAI's Codex:
bashcd your-project codex mcp add mem-mcp -- uvx --from git+https://github.com/memovai/memov.git mem-mcp-launcher stdio $(pwd)
VS Code
Create
in your project root:code.vscode/mcp.json
json{ "servers": { "mem-mcp": { "type": "stdio", "command": "uvx", "args": [ "--from", "git+https://github.com/memovai/memov.git", "mem-mcp-launcher", "stdio", "${workspaceFolder}" ] } } }
Reload VS Code to activate the MCP server.
Cursor
- Open Files > Preferences > Cursor Settings > MCP
- Add this configuration:
json{ "mcpServers": { "mem-mcp": { "command": "uvx", "args": [ "--from", "git+https://github.com/memovai/memov.git", "mem-mcp-launcher", "stdio", "${workspaceFolder}" ] } } }
Note: Some editors (like Antigravity) don't support
. Use the absolute path instead.code${workspaceFolder}
Claude Desktop
Edit
(Windows:code~/.config/claude/claude_desktop_config.json
):code%APPDATA%\Claude\claude_desktop_config.json
json{ "mcpServers": { "memov": { "command": "uvx", "args": [ "--from", "git+https://github.com/memovai/memov.git", "mem-mcp-launcher", "stdio", "." ] } } }
Enable RAG Mode (Optional)
For semantic search capabilities, install with the
extras:code[rag]
bashclaude mcp add mem-mcp --scope project -- uvx --from "git+https://github.com/memovai/memov.git[rag]" mem-mcp-launcher stdio $(pwd)
RAG mode enables natural language search across your session history.
Verify Installation
After connecting, verify MemoV is working:
- Start your AI coding tool
- Ask it to create or modify a file
- Run
to see the captured sessioncodemem history
You should see your prompt and the code changes recorded.
Troubleshooting
"uvx: command not found"
Reinstall uv and restart your terminal:
bashcurl -LsSf https://astral.sh/uv/install.sh | sh source ~/.bashrc # or ~/.zshrc
MCP Server Not Starting
- Verify uv is installed: code
which uvx - Test manual launch: code
uvx --from git+https://github.com/memovai/memov.git mem-mcp-launcher stdio . - Check your editor's MCP logs for errors
RAG Tools Not Available
If semantic search isn't working, reinstall with extras:
bashuvx --from "git+https://github.com/memovai/memov.git[rag]" mem-mcp-launcher
Next Steps
Now that MemoV is installed:
- Read about core concepts to understand snapshots and branches
- Explore the CLI reference for all available commands
- Launch the Web UI with code
mem ui - Check the architecture guide for technical details