Everything you need to observe, debug, and improve your OpenClaw agents.
Every run logged with status, model, provider, duration, token counts, and full event timelines. Filter by status, channel, model, or date range.
A three-pane editor for reviewing, tagging, pinning, and archiving agent memories with full provenance tracking.
View and edit your agent's structured identity: purpose, principles, tone, dos, donts, tool policy, escalation rules, and more.
Understand token and tool spend as runs happen. Stat cards, breakdowns by model and provider, and a full cost ledger.
Every tool your agent uses is auto-discovered and tracked with call counts, error rates, recent failures, and status.
Connection status, telemetry mode, ingestion key management, and security settings in one place.
Create an agent, pair the connector, and watch everything populate in real time.
