Real-time observability dashboard for OpenClaw AI agents
ClawMetry is a free, open-source observability dashboard for OpenClaw AI agents. Think Grafana, but purpose-built for AI. One command install (pip install clawmetry), zero config. Monitor token costs, sub-agent activity, cron jobs, memory changes, and session history. All in real-time with a beautiful live flow visualization. Works on macOS, Linux, Windows, even Raspberry Pi
ClawMetry tracks token usage and costs per session, model, and tool in real-time. Users can set budget limits and alerts, detect anomalies in spending patterns, and identify agents consuming 3x more tokens than necessary—enabling cost optimization before invoices arrive.
ClawMetry displays live sub-agent spawn trees, detects runaway loops and hallucinations, and allows one-click termination of problematic agents. Budget-crossing alerts and anomaly detection flag issues before they spiral into multi-month cost increases.
ClawMetry provides live flow visualization showing message routing, tool calls, sub-agent spawning, and decision paths as they happen. Users see which websites agents visit, where they get stuck, and the exact sequence of actions taken—replacing guesswork with mission-control-level clarity.
Starter and Pro plans support unlimited nodes and multi-runtime monitoring (Claude Code, Codex, Cursor, Aider, Goose, opencode, Qwen, Hermes). Fleet view, cloud sync, and per-runtime health timelines let teams observe diverse agent stacks without fragmented tools.
ClawMetry logs every session with timeline, tool calls, costs, and full transcripts. Users can replay what happened last Tuesday, trace decision paths step-by-step, and perform error triage to resolve or mute known failures—enabling rapid incident investigation.