clawmetry

See your agent think. Zero-config observability & governance for 30 AI agent runtimes: Claude Code, OpenAI Codex, Hermes, OpenClaw & 26 more. Live token costs,

What is clawmetry?

See your agent think. Zero-config observability & governance for 30 AI agent runtimes: Claude Code, OpenAI Codex, Hermes, OpenClaw & 26 more. Live token costs, sessions, tool calls, crons.

Overview of clawmetry

What it is

ClawMetry reads the session files and logs that AI agent runtimes already write, and presents a timeline of tool calls, token usage, and cost data in a single dashboard. It works with 33 AI agent runtimes including Claude Code, OpenAI Codex, Hermes, OpenClaw, and others. No SDK, code change, or instrumentation is required. The tool auto-detects installed runtimes, reads them read-only, and changes nothing about how they run.

Key features

  • Sessions & transcripts: view what each agent did turn by turn with replay - Cost & tokens: breakdown per runtime, model, session, and day with anomaly flags - Flow: live diagram of messages moving through channels, models, and tools - Context blowout: window utilization sized per provider, compaction vs forced overflow - Health & logs: disk, memory, error rates, rate limits, live log stream - Alerts: budget caps, error spikes, agent-offline, routed to Slack, Discord, PagerDuty, Telegram, Email - Approvals: pause risky tool calls before they run and approve from a phone

Use cases

Monitor whether an agent is making progress or stuck in a loop. Identify repeated failures and context-window blowout across different runtimes. Track cost and token consumption per session or per day. Set budget caps and receive alerts for error spikes or offline agents. Approve or block risky tool calls before execution.

Getting started

bash pip install clawmetry && clawmetry Open http://localhost:8900. Zero configuration required. If no agents are installed on the machine, run clawmetry --sample to load three labelled synthetic sessions.

License and pricing

License: MIT. Pricing plans (as stated in the README):

  • Free: OpenClaw, NVIDIA NemoClaw, Goose, Qwen Code – full dashboard, local only – $0 - Starter: every other runtime, fleet view, cloud sync – $9 per node / month - Pro: Starter + approvals, tool-risk policies, evals, anomaly detection, cost optimizer, OTel export, tamper-evident audit log – $19 per node / month Annual plans and enterprise pricing are listed at clawmetry.com/pricing.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/vivekchandclawmetry.svg)](https://yunodb.com/tools/vivekchandclawmetry?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/vivekchandclawmetry?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/vivekchandclawmetry.svg" alt="Featured on YunoDB" height="20"></a>