On this page you will connect MMW to Claude Code, verify the connection and, if you want, turn on session history upload.
What you get
Section titled “What you get”- The
remember,search,forget,validate_memoryandgateway_calltools in every Claude Code session. - The read-only
mmw_sync_statustool, when you connect through the local agent. - Claude Code session history upload, only after you explicitly consent in a terminal.
- The same memory that Cursor, Codex and claude.ai see with the same key.
Prerequisites
Section titled “Prerequisites”- An MMW API key (starts with
mmw_); see the Quickstart. - Claude Code with the
claudecommand in your terminal. - For the local agent: Python 3.11+ and uv (the
uvxcommand).
-
Run in a terminal:
Terminal window claude mcp add mmw -s user \-e MMW_API_KEY=mmw_your_key \-e MMW_ENDPOINT=https://mcp.mmwhub.tech \-- uvx --from https://app.mmwhub.tech/downloads/mmw-agent/mmw_agent-0.1.4-py3-none-any.whl mmw-agentThe
-s userflag makes MMW available in every project. Without it the server is added for the current directory only. -
Restart Claude Code if it was open.
-
Verify the connection (see below).
The command writes an mcpServers.mmw block to ~/.claude.json. You can also add it by hand:
{ "mcpServers": { "mmw": { "command": "uvx", "args": ["--from", "https://app.mmwhub.tech/downloads/mmw-agent/mmw_agent-0.1.4-py3-none-any.whl", "mmw-agent"], "env": { "MMW_API_KEY": "mmw_your_key", "MMW_ENDPOINT": "https://mcp.mmwhub.tech" } } }}Nothing to install, but no session history and no mmw_sync_status.
-
Run in a terminal:
Terminal window claude mcp add --transport http mmw https://mcp.mmwhub.tech/mcp -s user \--header "Authorization: Bearer mmw_your_key" -
Restart Claude Code and verify the connection.
Verify
Section titled “Verify”-
In a terminal run
claude mcp list: themmwserver should be listed as connected. -
Inside a Claude Code session type
/mcp: themmwserver showsremember,search,forget,validate_memoryandgateway_call, plusmmw_sync_statuswhen connected through the agent. -
Ask:
Remember in MMW: staging is deployed with the systemd unit app-staging. Fact key: deploy-staging.Claude Code calls
rememberand reports the record ID with theunverifiedstatus. -
Open a new session and ask:
How do we deploy staging? Check MMW.Expected answer: the assistant calls
searchand replies that staging is deployed withapp-staging, citing the MMW record.
You can check the agent on its own, without Claude Code:
MMW_API_KEY=mmw_your_key MMW_ENDPOINT=https://mcp.mmwhub.tech \ uvx --from https://app.mmwhub.tech/downloads/mmw-agent/mmw_agent-0.1.4-py3-none-any.whl mmw-agent statusSession history (optional)
Section titled “Session history (optional)”The agent can upload your Claude Code history to MMW, but only after a person confirms consent in an interactive terminal. A model or a script cannot grant consent.
mmw-agent consent # shows how many sessions it found and asks for permissionmmw-agent revoke # withdraw consent: uploads stop immediatelyIf mmw-agent is not installed globally, run it as uvx --from <wheel URL> mmw-agent consent with the same MMW_API_KEY and MMW_ENDPOINT.
User and assistant messages, tool calls and tool results are uploaded; model reasoning, system prompts, file snapshots and paths are not. Secrets are removed on your computer before sending and once more on the server. See Session history.
Common problems
Section titled “Common problems”For other errors see Errors and Agent troubleshooting.