| name | lobu |
| description | Build, run, and maintain Lobu agent projects, including lobu.toml, prompt files, local skills, evals, providers, connections, and Lobu memory workflows. |
Lobu
Use this skill when the user is working on a Lobu project or wants to scaffold, run, validate, evaluate, or connect one. Also use it when the user wants persistent Lobu memory, MCP client setup, OpenClaw memory plugin configuration, knowledge search/save workflows, watchers, or browser-authenticated connectors.
Core Model
- Lobu is the agent framework, runtime, deployment layer, and memory surface.
- Keep framework configuration in
lobu.toml.
- Keep agent identity and behavior in
IDENTITY.md, SOUL.md, and USER.md.
- Keep reusable capability bundles in
skills/<name>/SKILL.md or agents/<agent>/skills/<name>/SKILL.md.
- Use
lobu login for CLI authentication. Do not use a separate memory login command.
- Use
lobu memory ... for memory operations, MCP client wiring, seeding, direct tool calls, and browser-auth capture.
Project Checklist
- Read
lobu.toml first.
- Read the active agent files under
agents/<id>/.
- Check local skills under
skills/ and agents/<id>/skills/.
- Use
lobu validate after config changes.
- Use
lobu eval when prompt or behavior changes.
Common Commands
npx @lobu/cli@latest init my-agent
npx @lobu/cli@latest run
npx @lobu/cli@latest validate
npx @lobu/cli@latest eval
npx @lobu/cli@latest login
Memory Defaults
Your long-term memory is powered by Owletto. Do NOT use local files (memory/, MEMORY.md) for memory.
- Owletto automatically recalls relevant memories when you receive a message.
- To save something, call save_knowledge with the content and an appropriate semantic_type.
- To search, call search_knowledge. Results include view_url links to the web interface.
- NEVER construct Owletto URLs yourself. When the user asks for a link, call search_knowledge to get the correct view_url.
- When the user says "remember this", save it to Owletto immediately.
Lobu Memory
Configure project-scoped memory in lobu.toml:
[memory.owletto]
enabled = true
org = "my-org"
name = "My workspace"
models = "./models"
data = "./data"
Then seed or operate the memory workspace with:
lobu login
lobu memory org set <org-slug>
lobu memory health --org <org-slug>
lobu memory seed --org <org-slug>
lobu memory run search_knowledge '{"query":"Acme"}' --org <org-slug>
Use search_knowledge first when the user asks about a specific entity or workspace memory. Use save_knowledge to persist durable memory. To update existing knowledge, search first, then save with supersedes_event_id so the old row is tombstoned rather than deleted.
MCP Client Setup
Use the actual MCP URL for the user's runtime. Never hardcode a hosted URL unless the user explicitly asks for that instance.
Common setup commands:
claude mcp add --transport http lobu <mcp-url>
codex mcp add lobu --url <mcp-url>
gemini mcp add --transport http lobu <mcp-url>
lobu memory init --url <mcp-url>
For ChatGPT, Claude Desktop, Cursor, and other browser-managed clients, paste the MCP URL into the client's MCP/connector settings and complete OAuth in the browser.
OpenClaw Memory Plugin
For OpenClaw, install the plugin and let the Lobu CLI write plugin config:
openclaw plugins install @lobu/openclaw-plugin
lobu login
lobu memory configure --url <mcp-url> --org <org-slug>
lobu memory health --url <mcp-url> --org <org-slug>
lobu memory configure writes a token command that uses lobu token --raw, so OpenClaw reuses the top-level Lobu login.
Browser-Authenticated Connectors
For connectors that need cookies from a local browser session:
lobu memory browser-auth --connector <key> --auth-profile-slug <slug>
lobu memory browser-auth --connector <key> --auth-profile-slug <slug> --check
Use --chrome-profile, --launch-cdp, and --dedicated-profile only when the user needs a specific browser profile or dedicated remote-debugging profile.
Tool Discipline
- Search before create to avoid duplicate entities.
- Never fabricate Lobu memory links. If a tool returns a view URL, use that URL.
- Use canonical MCP tool names only.
- Prefer read-only operations before mutations when validating connectivity.
events is append-only: never delete rows directly; use tombstone/supersede flows.