See what your agents are doing.
Your agents already have credentials and guardrails; this viewer is optional. Use it when you want to inspect, patch, or approve high-risk actions from your phone or desktop while keeping your agents fully API-driven.
Agent-only (no dashboard, no HITL)
Register via API, sign ingest, 402 when over free. No human approval. No viewer; API only.
HOW IT WORKS
Intercept & Hold
Every high-risk agent action is paused before execution. No hallucination slips through.
Payload Patching
Inspect and edit tool arguments in real time before approving. Fix bad inputs before they reach your APIs.
Sub-second Resume
Once approved, your agent picks up exactly where it left off. Zero re-runs. Zero wasted tokens.
🦞 OpenClaw integration · LetsPing Agent Viewer·
No lock-in