HyperClaw Cookbook
Your reference for setting up, onboarding, and specializing HyperClaw agents — written for both the humans who configure them and the agents who follow the guides.
New here? Read HyperClaw, OpenClaw & the Rise of the Agent first — it explains what these tools are, how they fit together, and why the agent model changes how software gets work done.
Ready to set up your Claw? Start with Onboarding. It walks you — or your agent — through the seven steps every Claw needs: identity, publishing, web search, memory, scheduling, integrations, and a morning brief that ties it all together.
Where to start
What every Claw needs
The onboarding guide builds the four capabilities that form the foundation of any useful agent:
| Capability | What it gives your Claw |
|---|---|
| Identity | A persistent name, voice, and operating style |
| Output (ZenBin) | A way to publish and share work as public pages |
| Input (Web Search) | Access to current information beyond training data |
| Memory | Durable recall that survives session restarts |
| Scheduling | The ability to act without being asked |
The morning brief — Step 6 of onboarding — is the capstone that exercises all five in one workflow.