Choose your setup

Which version is right for you?

OSS / Self-Hosted

Free & open source

✓ Free, forever
✓ Open source on GitHub
✓ Runs entirely local. One machine.
✓ Your data stays on your machine
✓ No account required
– Single node only
– No remote access

Paid self-hosted licenses unlock all runtimes and paid features locally. See pricing.

Get started free

LIVE

☁️

ClawMetry Cloud

Managed fleet

✓ Managed & always on
✓ End-to-end encrypted
✓ Multi-node fleet (10+ Mac Minis)
✓ Access dashboard from anywhere
✓ Team access & sharing
✓ Zero plaintext on our servers
✓ One view for all nodes

Get Started

Architecture

No proxy. No code changes.

ClawMetry reads the session files and logs your agent runtimes already write - OpenClaw, Claude Code, Codex, Cursor + 13 more. Nothing sits between your agents and the models.

🤖

Your AI agents

OpenClaw, Claude Code, Codex
+ 14 more. Nothing changes.

reads files

ClawMetry

Parses logs + sessions.
Serves dashboard.

shows you

📊

Your browser

localhost:8900
Live dashboard

Runs locally on the same machine as your agents. Your data lives in a local database on that machine.

Supported runtimes

17 runtimes. One dashboard.

ClawMetry auto-detects every supported runtime on your machine and shows them side by side - same tabs, same live feed, same cost attribution. No SDK, no code changes.

OpenClaw NVIDIA NemoClaw NanoClaw PicoClaw Claude Code Codex Cursor Aider Goose OpenCode Qwen Hermes Pi Deep Agents n8n Antigravity GitHub Copilot

OpenClaw and NVIDIA NemoClaw are free on every plan.

Agent Observability

X-ray vision into your agent's runtime

See exactly what happens at every layer - from incoming messages to LLM calls to tool execution. Gateway runtimes like OpenClaw map to all three layers; CLI runtimes like Claude Code and Codex are observed at the runtime layer, straight from their session files.

Layer 1: Connectors

📲

3 groups, 1 DM

💬

1 group

🎮

Discord

2 channels

⌨️

CLI

main session

🔌

routes messages

Layer 2: Gateway Controller

Main

TG:group1

TG:group2

WA:chat

DC:chan1

Heartbeat

Each channel/group gets its own isolated session with full transcript

Cron Manager

Memory Mgmt

Session DB

builds context + calls LLM

Layer 3: Agent Runtime

Environment

Claude Code

GCP

exe.dev

Providers

Opus 4.6

GPT 5.4

Sonnet 4.6

Tools

web_search

read_file

exec

Skills

weather

gh-issues

tmux

Brain Tab

Channel + Skill Badges

Every event tagged with which channel it came from and which skill triggered it. Filter by Telegram, WhatsApp, Discord, or CLI. Click any turn to see the full execution trace.

Context Tab

LLM Context Inspector

See exactly what context the LLM receives: system prompt breakdown, bootstrap files (SOUL.md, AGENTS.md), tool schemas, conversation history, and compaction events.

Skills Tab

Skills File Browser

IDE-like explorer for all installed skills. Click any skill to browse SKILL.md, scripts, and references. See fidelity stats: which skills are actually being used vs dead weight.

Per-Turn Execution Timeline

          18:12:50
          USER
          Telegram
          "deploy to prod"
        

          5 steps
          2 LLM
          2 tools
          5.6s
        

            18:12:52
            EXEC
            git status
          

            18:12:52
            READ
            deploy.sh
          

            18:12:53
            THINK
            Checking deployment config...
          

            18:12:55
            AGENT
            Deployed successfully to prod
          

LLM Context Composition

What the model actually sees each turn

## Tooling

3.0K

## Skills headers

1.5K

Bootstrap: SOUL.md

750

Bootstrap: MEMORY.md

1.0K

Tool schemas (JSON)

7.0K

Conversation history

38.5K

Total: 51.7K / 200K tokens (26%) Compaction threshold: ~160K

ClawMetry Cloud

Monitor a fleet of Mac Minis from anywhere

Each node end-to-end encrypts your agents' conversations and activity before sending; only you hold the key, so the cloud stores ciphertext it cannot read. Simple totals like cost and health sync securely, and you see every machine in one dashboard, wherever you are.

🖥️ Mac Mini 1 🔒

🖥️ Mac Mini 2 🔒

🖥️ Mac Mini 3 🔒

🖥️ Mac Mini 4 🔒

🖥️ Mac Mini 5 🔒

🖥️ Mac Mini 10+ 🔒

E2E encrypted

☁️

ClawMetry Cloud

Zero plaintext
on our servers

anywhere

🌐

Your dashboard

Any browser,
anywhere

🔐

E2E Encrypted

Data is encrypted on-device before it leaves your Mac Mini

🚫

Zero plaintext on our servers

ClawMetry Cloud only ever sees ciphertext. Only you can decrypt.

🗺️

Any node, one view

All your Mac Minis in one dashboard. Switch between nodes instantly.

Setup

Running in under a minute

Two commands. That's it.

terminal

Install with pip

Works wherever Python 3.8+ is installed. Same machine as your agents.

Run `clawmetry`

Auto-discovers the AI agent runtimes on your machine. Opens the dashboard at localhost:8900.

That's it

No account. No API key. No config file. Open your browser and start watching.

What you see

Everything in one dashboard

Six tabs. Each one answers a different question about what your agents are doing.

🤖

Main Agent

Live feed of every tool call your agents make - across OpenClaw, Claude Code, Codex and every other runtime. Name, what it did, how long it took, updated in real time.

Live

🔀

Sub Agents

7-day history of every sub-agent run. See what task it was given, when it started, when it finished, and what it did along the way.

7-day history

💰

Token flow

Real-time token usage broken down by model and session. Know exactly which sessions are burning through your budget.

Live

⏰

Cron jobs

Every scheduled task in your agent setup - last run time, status, and run history. Catch a missed cron before it matters.

Live

🩺

System health

Service status for your agent gateway and connected components at a glance.

Live

📜

Session history

Browse past sessions with a full timeline of what happened - tool calls, costs, duration. Useful for debugging anything that went wrong.

Searchable

Under the hood

Reads each runtime's own files

ClawMetry doesn't intercept traffic or modify your setup. Every runtime already writes its activity to disk - OpenClaw its workspace, Claude Code and Codex their session transcripts, Cursor its local state. ClawMetry reads those native files and presents them as one live dashboard.

This means zero overhead on your agents and zero risk of ClawMetry breaking anything.

📁

Session transcripts Every tool call, model response, and agent turn, timestamped - parsed from each runtime's native format

⏰

Cron job registry Scheduled tasks, run history, and last-fired timestamps

🧠

Workspace files Memory files, SOUL.md, AGENTS.md - whatever your agent reads and writes

🔌

Gateway status OpenClaw gateway health, active sessions, connected channels

~/.openclaw/workspace

📁 sessions/ ← session transcripts

📁 memory/ ← memory files

📄 MEMORY.md ← long-term memory

📄 SOUL.md ← agent persona

📄 HEARTBEAT.md

~/.openclaw/cron

📄 jobs.json ← cron jobs

~/.openclaw/agents

📁 main/ ← main agent sessions

📁 sub-*/ ← sub-agent runs

Other runtimes (auto-detected)

📁 ~/.claude/projects/ ← Claude Code

📁 ~/.codex/sessions/ ← Codex

📄 …/Cursor/…/state.vscdb ← Cursor

📄 + 13 more runtimes

ClawMetry reads these. Your agents keep writing them. Nothing changes.

One command.
Full visibility.

Which version is right for you?

No proxy. No code changes.

17 runtimes. One dashboard.

X-ray vision into your agent's runtime

Monitor a fleet of Mac Minis from anywhere

Running in under a minute

Install with pip

Run `clawmetry`

That's it

Everything in one dashboard

Main Agent

Sub Agents

Token flow

Cron jobs

System health

Session history

Reads each runtime's own files

See what your agents are actually doing

Stay in the loop

One command.Full visibility.

Which version is right for you?

No proxy. No code changes.

17 runtimes. One dashboard.

X-ray vision into your agent's runtime

Monitor a fleet of Mac Minis from anywhere

Running in under a minute

Install with pip

Run clawmetry

That's it

Everything in one dashboard

Main Agent

Sub Agents

Token flow

Cron jobs

System health

Session history

Reads each runtime's own files

See what your agents are actually doing

Stay in the loop

One command.
Full visibility.

Run `clawmetry`