# ClawMetry

> Real-time observability dashboard for OpenClaw AI agents with live flow visualization, token tracking, and sub-agent monitoring.

ClawMetry is an open-source, real-time observability dashboard designed specifically for OpenClaw AI agents. It provides complete visibility into what your agents and sub-agents are doing at any moment, including file reads, command executions, tool calls, and reasoning processes. The tool runs entirely locally with zero configuration required and is free forever.

- **Live Flow Visualization** provides an animated SVG architecture diagram that lights up in real-time as your agent processes messages, with color-coded indicators for different event types including messages, LLM requests, tool calls, responses, and errors.

- **Sub-Agent Monitoring** shows every step when your agent spawns sub-agents, displaying files being read, commands being run, tools being called, and what the sub-agent is thinking, available as summary, narrative, or full logs.

- **Token & Cost Tracking** monitors token usage and costs per session, per model, and per tool with bar charts showing daily, weekly, and monthly totals, helping you understand spending before invoices arrive.

- **Cron Job Dashboard** monitors scheduled jobs with status indicators, schedule expressions, last/next run times, and duration with alerts when cron jobs fail.

- **Session History & Transcripts** logs every session with timeline, tool calls, and cost, featuring a transcript viewer that shows conversations in a chat-bubble UI with color-coded roles.

- **Memory File Browser** allows browsing and tracking changes to your agent's memory files including SOUL.md, MEMORY.md, AGENTS.md, and daily notes.

- **System Health Dashboard** provides at-a-glance monitoring of service uptime, disk usage, active sub-agents, and cron job status.

To get started, install via pip with `pip install clawmetry` and run `clawmetry` to open the dashboard at localhost:8900. The tool auto-detects your OpenClaw workspace with zero configuration. ClawMetry runs on macOS, Linux, Windows WSL, and even Raspberry Pi, requiring only Python 3.8+ and Flask.

## Features
- Live flow visualization with animated SVG architecture diagram
- Real-time sub-agent monitoring
- Token and cost tracking per session, model, and tool
- Cron job dashboard with status indicators
- Session history and transcript viewer
- Memory file browser for agent memory files
- System health monitoring
- Mission Control integration with OpenClaw task system
- Zero configuration auto-detection
- Self-hosted and runs entirely locally

## Integrations
OpenClaw, OpenTelemetry

## Platforms
WINDOWS, MACOS, LINUX, WEB, API

## Pricing
Open Source

## Links
- Website: https://clawmetry.com
- Documentation: https://github.com/vivekchand/clawmetry/blob/main/README.md
- Repository: https://github.com/vivekchand/clawmetry
- EveryDev.ai: https://www.everydev.ai/tools/clawmetry