ClawMetry for OpenClaw
Real-time observability dashboard for OpenClaw AI agents
634 followers
Real-time observability dashboard for OpenClaw AI agents
634 followers
ClawMetry is a free, open-source observability dashboard for OpenClaw AI agents. Think Grafana, but purpose-built for AI. One command install (pip install clawmetry), zero config. Monitor token costs, sub-agent activity, cron jobs, memory changes, and session history. All in real-time with a beautiful live flow visualization. Works on macOS, Linux, Windows, even Raspberry Pi
This is the 2nd launch from ClawMetry for OpenClaw. View more
ClawMetry Cloud
Launched this week
Your OpenClaw agent is running, but do you know what it's doing or how much it's costing? ClawMetry Cloud gives you live flow visualization, token costs, memory state, and sub-agent activity from any browser or Mac app. Two commands to connect: pip install clawmetry, then clawmetry connect. E2E encrypted, your data never touches our servers unencrypted. Open source locally, $5/node/month for cloud. 7-day free trial.
Free Options
Launch Team / Built With
ClawMetry for OpenClaw
Hey Product Hunt! 👋
I'm Vivek, the maker of ClawMetry. A few weeks ago, I launched the open-source version here and you gave an incredible #5 Product of the Day. Thank you!
But after talking to dozens of users, one thing kept coming up: "I love the dashboard, but I can't see it when I'm away from my desk."
That's the problem ClawMetry Cloud solves. It syncs your AI agent metrics to the cloud so you can check on your agents from any browser, anywhere. The catch? We built it with end-to-end encryption, your data is encrypted before it ever leaves your machine. Even we can't read it.
What's new in Cloud:
• 🔒 E2E encrypted sync (zero-knowledge architecture)
• 🌍 Browser access from anywhere (https://app.clawmetry.com/)
• 📊 Multi-node monitoring (all your agents, one dashboard)
• 🍎 Native Mac app
Some context on traction:
• 75K+ PyPI downloads
• Built for the OpenClaw ecosystem (316K GitHub stars)
• 7-day free trial, then $5/node/month
I'd love to hear: What's the biggest pain point you have monitoring your AI agents in production?
Happy to answer any questions! 🙏
okay this is cool. “can’t see it when i’m away from my desk” is such a real product moment lol. a lot of tools feel fine till you actually need to check something quickly from outside, then that gap becomes super obvious. also love that you didn’t ignore the privacy side while adding cloud sync.
curious, what’s the first thing people usually open clawmetry cloud to check when they’re away?
ClawMetry for OpenClaw
@nayan_surya98 Ha, exactly. The gap only becomes obvious when you're at a resort on vacation and wondering what your team of 5 Mac Minis are doing back home. The first thing most people check: the Flow tab to see if the agents are still running, and the token cost for the last session.
But the real unlock is the Brain activity tab. After you send "can you fix this issue?", you can watch in real time: what it's thinking, which files it's reading, which tools it's calling, whether it went down the right path or got confused halfway through. "Did anything break, how much did it cost me, and did it actually do what I asked?" sums up the away-from-desk experience pretty well.
This is a smart extension of the original product. Local observability makes sense at first, but the moment agents start doing real work, being able to check on them remotely becomes much more important. The end-to-end encryption angle makes this a lot more convincing too, because monitoring data can get sensitive very quickly.
Curious, what tends to matter most to users once they go remote, uptime visibility, cost tracking, or understanding agent behavior over time?
ClawMetry for OpenClaw
@akshay_kumar_hireid Akshay, great question. From what I've seen, cost tracking wins hands down once agents go remote. The "why did this run cost $12?" question comes up almost immediately. Uptime visibility is second, especially for people running crons overnight. Understanding behavior over time is the longer-term value but takes a few weeks of data before it clicks. The E2E encryption piece matters more than people expect too - once agents have access to email, calendar, and files, the monitoring data itself gets sensitive fast.
Okan
Having E2E encryption built right into the monitoring layer is a massive relief for anyone running OpenClaw agents on untrusted networks. I would definitely use this to keep tabs on my remote scraping nodes without having to tunnel into each server manually. I am really curious to hear how you handle the initial key exchange process when provisioning a new agent from the command line.
ClawMetry for OpenClaw
@y_taka thanks for the feedback! Regarding the E2E encryption, When you run clawmetry connect, the CLI generates a random AES-256 encryption key locally on your machine. This key never leaves your device. The CLI then authenticates with our API using your cm_ API key, and starts pushing encrypted snapshots. The encryption key is stored in ~/.clawmetry/config.json alongside your API key. When you open the dashboard in the browser, you enter (or paste) that same key once, and it's stored in localStorage. All decryption happens client-side in the browser via Web Crypto API. Our servers only ever see encrypted blobs, never the key, never the plaintext.
Solid tool real-time observability for AI agents is super practical, especially for devs managing multiple sub-agents.
Could land stronger with the key benefit up front Monitor your Open Claw agents live costs, activity, and memory without any setup.
ClawMetry for OpenClaw
@allu__kurashi Thanks for the feedback, I've updated the messaging!
co.dev
congrats on the launch!