Clawmetry: Free Open-Source Real-Time Dashboard for OpenClaw AI Agents

If you're running OpenClaw agents, you already know the problem: your agent says "spawning sub-agent" and then you just... hope it works. You have no visibility into what it's actually doing — which files it's reading, which tools it's calling, how many tokens it's burning, or whether it's even on the right track.
Clawmetry fixes that. It's a real-time observability dashboard built specifically for OpenClaw, open source, MIT licensed, and free forever.
What Clawmetry Does
Clawmetry gives you a single screen showing everything happening across your OpenClaw agents in real time. It hooks into OpenClaw's runtime and surfaces information that was previously invisible:
- Sub-agent visibility — See every step of every sub-agent: what files it's reading, what commands it's running, what tools it's calling, and what it's thinking. Summary, narrative, and full logs available.
- Live flow graph — Channels, gateway, models, tools, and nodes all displayed as a live graph that updates as your agents work.
- Token and cost tracking — Tokens in, tokens out, cache hits, response times, and cost per call. Know what you're spending in real time.
- System health — Cron jobs, service uptime, disk usage, and active sub-agents at a glance.
- Session history — Every session logged with a full timeline, tool calls, and cost. Go back and see exactly what happened.
- Mission Control integration — Hooks into OpenClaw's task system so you can see which agents are working on which tasks.
- Cost breakdown — Per-session, per-model, per-tool. Know what you're spending before the invoice shows up.
One Command Install
Getting started is deliberately frictionless. On macOS, Linux, or Windows (WSL), the entire install is a single command:
curl -fsSL https://clawmetry.com/install.sh | bash
The installer detects your OS and handles the rest. If you'd rather not self-host, Clawmetry also offers managed instances and free onboarding.
Works Everywhere OpenClaw Runs
Clawmetry is designed to run wherever OpenClaw runs — laptop, server, Raspberry Pi, or any cloud provider including AWS, GCP, Hetzner, Digital Ocean, and Hostinger.
Why This Matters
As OpenClaw gains traction as an agentic AI framework, the need for observability tools has grown rapidly. Multi-step agents that spawn sub-agents and call multiple tools are inherently hard to debug — you need to see inside them while they run, not just read logs after the fact.
Clawmetry fills this exact gap. It was built because the developer needed it and there was nothing else that worked. Now you don't have to build it yourself.
The Bottom Line
If you're building with OpenClaw, Clawmetry is the missing piece. Real-time visibility into every sub-agent, tool call, token cost, and session log. Open source, MIT, free forever. Installs in one command.
GitHub: github.com/vivekchand/clawmetry