How do I monitor Google Antigravity sessions?

Install ClawMetry (pip install clawmetry). The daemon auto-detects Antigravity and renders every session, tool call, model, and Gemini token in one dashboard at http://localhost:8900. No SDK.

Does ClawMetry catch background Antigravity generations that keep burning tokens?

Yes. Antigravity's background-generation mode keeps a session running after you close the tab; ClawMetry watches the brain log for continued write activity and flags a session whose token burn continues after the last user interaction. The kill switch (Pro) stops the generation before the next call.

Does ClawMetry track Gemini thinking-token cost?

Yes. Gemini's thinking-mode bills thinking tokens on top of output; ClawMetry prices thinking, output, and prompt tokens separately on the Cost tab so you see the real bill, not the model's stated output cost.

Does ClawMetry work with all four Antigravity surfaces (CLI, IDE, brain, jetski)?

Yes. ClawMetry auto-detects all four Antigravity surfaces and labels every session with the surface it came from, so CLI, IDE, brain, and jetski activity stay separable in the dashboard.

How much does ClawMetry cost for Antigravity?

Free forever on your own machine. ClawMetry Cloud starts at $9 per node per month (Starter) or $19 per node per month (Pro, which adds policy gates, evals, and background-generation kill switches). Antigravity is one of 18 paid runtimes on the Cloud tier.

Antigravity × ClawMetry

Observability and cost tracking for Google Antigravity

See every Antigravity session, tool call, and Gemini token as it happens. ClawMetry detects Antigravity automatically and renders sessions, tokens, cost, and background-generation burn in one dashboard.

Supported runtime Zero config, auto-detected Read-only Free trial Official Antigravity site ↗

📦 631k+ installs 🌍 128+ countries ⭐ 392 on GitHub 🏆 #5 Product Hunt

What ClawMetry sees on Antigravity

Once ClawMetry is running, your Antigravity activity shows up as:

Sessions: started at, last activity, model, message count, per-session token totals.
Live activity: every tool call as it lands, with args and result. Turn-level tokens split.
Per-turn tokens: Gemini-shaped per-generation token split (prompt tokens, candidates tokens, cached content tokens, thinking tokens) plus derived cost.
Cost per session, per model, per hour: cache-aware where the model supports caching. Compare Antigravity spend against the other 19 runtimes ClawMetry observes.
Background generations that keep burning tokens after you close the tab: the Antigravity-specific signal ClawMetry surfaces that generic dashboards do not.
Approval + kill switch: watch for token-maxing signals (message repeat, no-progress heuristic, cost-per-hour threshold), pause or kill the session before the next call.

Enable it in about two minutes

1. Install ClawMetry

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.

2. Run the daemon

clawmetry

The daemon auto-detects Antigravity and starts reading immediately. No env var, no config, no runtime restart.

3. Open the dashboard

Open http://localhost:8900. Switch the runtime chip to Antigravity.

Start the free trial →

The one thing this catches that Antigravity alone does not

Left alone, Antigravity's failure mode is a background generation left running overnight that quietly bills you for 10 hours of Gemini Pro reasoning. ClawMetry watches for it: cost per hour on the Antigravity runtime, message-repeat detection, and per-session token-growth curves. Set a threshold, and the approval queue pauses before the next call. Kill switch (Pro) stops it.