google-ai-usage-monitor

What This Skill Does

The google-ai-usage-monitor is a specialized OpenClaw skill designed to automate the oversight of your Google AI Studio resource utilization. By leveraging browser automation, it navigates the Google AI Studio usage dashboard, extracts real-time metrics including Requests Per Minute (RPM), Tokens Per Minute (TPM), and Requests Per Day (RPD), and compares these against your specific quota limits. The skill identifies if you are approaching critical thresholds and provides structured, actionable alerts directly to your team via Discord or other communication channels.

Installation

To install this skill, use the OpenClaw CLI in your terminal: clawhub install openclaw/skills/skills/dongsjoa-byte/google-ai-usage-monitor Ensure you have configured your project settings and provided the necessary browser environment access within your OpenClaw agent profile to allow for the navigation of the Google AI Studio dashboard.

Use Cases

This skill is indispensable for development teams scaling applications built on Gemini models. It is ideal for:

1. Preventing unexpected downtime caused by hitting rate limits (429 errors).
2. Providing project stakeholders with automated weekly health reports on API consumption.
3. Optimizing costs by identifying if you are consistently hitting tier-based limits, justifying a move to a higher payment tier.

Example Prompts

1. "OpenClaw, run a usage check on my current Google AI Studio project and post the report to the #dev-alerts channel."
2. "Check the current quota usage for the 'Production-Gemini' project and tell me if any model is exceeding the 80% usage threshold."
3. "Set up a recurring daily monitoring task for Google AI usage and alert me only if usage crosses the 50% mark."

Tips & Limitations

• Efficiency: The dashboard loads data asynchronously; ensure your internet connection is stable so the browser agent has sufficient time to render the tables before extraction.
• Authentication: Ensure the profile running the skill is logged into the specific GCP account associated with your project; otherwise, the page may redirect to a login screen.
• Limitations: Actual limits are dynamic and vary by model type. If Google updates the UI structure of AI Studio, the parsing logic may require a minor update. Always verify manual values if the dashboard displays an 'unknown' status.