Guide for FDEs

For FDEs and AI delivery builders, the hard part was never the demo - it is what happens after handover: can the client maintain it? Can you explain the data boundary? Who fixes it when it breaks? MotiClaw turns all three into product capabilities.

Before delivery: qualify the scenario

• Sensitive data (client lists, contracts, quotes) → the local-first architecture answers compliance concerns directly.
• No in-house tech team → one-click install, repair, and update mean no on-site ops.
• The client wants working workflows, not model APIs → the pre-configured agent library is a ready workflow catalog.

During delivery: build the workbench

1. Install MotiClaw on the client's device and configure models and gateway (their own gateway works too).
2. Claim and customize agents from the library for the client's business: role, skills, channels.
3. Connect the client's existing channels such as Feishu so agents live inside real workflows.

After delivery: verifiable and maintainable

• Verifiable boundary: data and agents stay on the client's device; only channel and model calls go out.
• Client-maintainable: Repair fixes most issues in one click; updates are one click too.
• Explainable cost: token usage and cost trends are visualized so clients understand what they pay for.