Your Grok Bot agents run on xAI's cloud VMs, where you cannot watch them — but they can ask your machine to run commands, and can hold standing permission to keep doing it. ClawMetry detects Grok Bot automatically, renders every conversation, and shows exactly what each bot is allowed to do locally.
Grok Bot is one of them; the other 27 are listed below. Free on every plan: OpenClaw, NVIDIA NemoClaw, Goose. Paid on Cloud Pro or Self-Hosted Pro: the rest.
Once ClawMetry is running, your Grok Bot activity shows up as:
Pick the line that matches your OS. Windows users: paste the install.cmd line, not the shell one.
# macOS or Linux
curl -fsSL https://clawmetry.com/install.sh | bash
# Windows (cmd.exe or PowerShell)
curl -fsSL https://clawmetry.com/install.cmd -o install.cmd && install.cmd
Already have Python? pip works too; onboard is the interactive first-run wizard.
# Any OS with Python 3.9+
pip install clawmetry && clawmetry onboard
Or skip the terminal entirely and grab the desktop app: macOS (.dmg) · Windows (.exe) · Linux (.AppImage). It installs and updates ClawMetry for you.
clawmetry
The daemon auto-detects Grok Bot and starts reading immediately. No env var, no config, no restart of the desktop app.
Open http://localhost:8900. Switch the runtime chip to Grok Bot.
Start the free trial →A bot asks once to run a command on your laptop, you click “always”, and the prompt never appears again. The grant is real, it persists, and nothing in the Grok Bot interface keeps a running list of what you have handed out. ClawMetry does: every permission ask, what it wanted to run, which bot asked, and which grants are standing rather than one-off — next to whether the local-exec helper is running right now.
That is the question worth answering for an agent you cannot watch. It is not spend, because spend is not knowable from here.
Engineers and founders from OpenAI, Google, PostHog and more have starred ClawMetry on GitHub. One of them helped build it.