Connect OpenClaw

Give every OpenClaw agent the same permission-aware company context used by Claude, ChatGPT, and your own agents.

Use mapped devices from OpenClaw

OpenClaw can request governed work that Lobu routes to an authorized device worker for local files, connector syncs, actions, or device Automations. Run lobu daemon on that machine once; a named context keeps one stable host identity, while supported Claude Code, Codex, and OpenCode sessions receive separate per-session identities for interactive delivery. See the device worker setup.

Lobu is open-source but we also have a managed cloud. If you'd like to try, sign in at app.lobu.ai, then point OpenClaw at https://app.lobu.ai/mcp as the MCP endpoint. The same URL works for every workspace. Pick one below to scope it to that example.

Paste this into OpenClaw and it will install Lobu for you.

Prompt for OpenClaw

Connect this OpenClaw session to Lobu, the shared company context layer for agents, over MCP: https://lobu.ai/mcp.

Onboard yourself:
- If you have a CLI: register the MCP endpoint above with whatever harness you run (claude mcp add / codex mcp add / pi / opencode), then read the Lobu skill and run search_sdk.
- If you can't touch a terminal: ask the user to add an MCP server named `Lobu` in OpenClaw pointing at https://lobu.ai/mcp, complete the Lobu sign-in, and pick the workspace — then continue once connected.

Authenticate when prompted, then:
Read the Lobu skill at https://app.lobu.ai/api/skill/lobu, then run search_sdk for the smallest SDK surface you need. Do not dump the whole workspace.
Search existing memory before answering or saving a duplicate. Use search_memory/save_memory for durable context, query_sdk for reads, run_sdk for writes. Preview consequential writes first and keep every returned source URL, identifier, and approval status exact — never invent one, never fabricate a returned URL or identifier.