A specialized utility for querying and monitoring usage statistics, token consumption, and remaining balances from any new-api deployment.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install new-api-usage
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install new-api-usage using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The new-api-usage skill provides developers and system administrators with a streamlined way to track resource consumption across AI model endpoints. By integrating with Openclaw Skills, this tool enables real-time auditing of API credits, usage logs, and model-specific performance data without exposing sensitive configuration files.
This utility is designed with a focus on security, requiring explicit credential input for every session to prevent accidental data leaks. It converts abstract quota numbers into readable dollar amounts, helping teams stay within budget while managing multiple AI model providers through a unified interface.
To use this tool within the Openclaw Skills framework, ensure your environment is configured for Python execution. No local configuration files are used, as credentials must be passed explicitly for security.
# Basic usage to check today's summary
uv run python scripts/query_usage.py --base-url https://api.example.com --key sk-your-key
The skill processes data from the new-api backend and organizes it into a structured format for easy analysis within Openclaw Skills:
| Field | Description |
|---|---|
| Quota | Raw numerical units (Conversion: 500,000 quota = $1) |
| Usage Records | Individual log entries containing token counts and timestamps |
| Model Aggregation | Grouped data showing consumption per specific model ID |
| Balance | Calculated remaining dollar value based on current quota |
Loading
A CLI-driven skill for deploying and managing high-performance vLLM-based Large Language Model services across multiple GPU servers.

An advanced AI-powered Reddit Voice of Customer research engine that transforms community discussions into professional business reports.

A specialized tool for extracting Amazon product data and automatically listing it on Shopify stores with multilingual translation.

Execute commands and manage files within a persistent Astra Docker container workspace.

Triggers local bash scripts when user queries begin with the Xiaoai assistant wake word.

A secure utility for broadcasting pre-signed hex transactions directly to the blockchain via OKX Web3 API without requiring local private keys.








































