ClawMetry sits alongside OpenClaw and continuously monitors every tool call, sub-agent, memory read, and token spent. When something goes wrong, you know immediately.
No proxy. No code changes. ClawMetry reads OpenClaw's output stream and maps it into a live dashboard with zero overhead on your agents.
From zero to full observability in under a minute.
One pip install, one command to connect. ClawMetry auto-discovers your OpenClaw gateway and starts listening immediately.
ClawMetry tails every event your agents produce. No instrumentation required. Works with your existing setup, unmodified.
When something looks wrong, you know first and can step in with one tap.
ClawMetry tracks the full lifecycle of every agent session - from first tool call to final output.
Every tool invocation: name, inputs, outputs, latency, and whether it succeeded or failed. Search and filter by tool type or session.
When sub-agents are spawned, what tasks they run, how long they take, and how they exit. No more blind spots in multi-agent chains.
Per-session, per-model, per-tool cost breakdown in real time. Know exactly which agents are burning your budget before the invoice lands.
Tool errors, retries, stuck loops, and silent failures surface immediately in the dashboard - and trigger alerts to your team.
Every memory file read and write, with before/after diffs. See exactly how your agents' long-term context changes over time.
Scheduled agent tasks with run history, success rates, and last-fired timestamps. Catch missed runs before they become incidents.
Not 3 hours later when you check your email. ClawMetry fires alerts the moment something goes wrong - to wherever your team already lives.
exec() with command: rm -rf /tmp/cache/*Observability without control is just logging. ClawMetry lets you pause, inspect, approve, and stop what your agents do, before it's too late to reverse.
Freeze a running agent mid-task with its state held, look at what it is doing, then resume right where it left off. Works on Claude Code, Codex, Goose, opencode, Aider, and OpenClaw.
When an agent is burning money or stuck in a loop, stop it in one tap from the dashboard or the desk device. You stay in control: ClawMetry surfaces the runaway, a human makes the call. Every stop is logged.
An agent that requests approval holds its work until you approve or deny, from the dashboard or the desk device.
See the exact tool call, inputs, and context the agent was about to use. Make the call with full information.
Flag specific tools or patterns that require human sign-off. Agents wait. You decide.
Free, open source, and running in under a minute. No account required.