Does it support multiple LLM providers?

Yes, it is designed to log metadata like model name and provider (e.g., OpenAI, Anthropic) alongside the usage data, allowing for multi-model cost analysis.

How are monthly invoices generated?

It utilizes Convex cron jobs to automatically aggregate monthly usage data and create invoice records for every active user at the end of a billing period.

Can I track usage per user?

Yes, the skill includes schema and query examples to attribute token consumption to specific user IDs for granular billing and individual analytics.

How does this skill calculate LLM costs?

It uses a custom usage handler that calculates costs based on input and output token counts multiplied by the specific provider's pricing rates defined in your Convex logic.

Can I set up automated alerts for high usage?

Absolutely, the skill provides implementation patterns for monitoring usage thresholds and triggering notifications when a user or agent exceeds a budget.

Convex Agents Usage Tracking

Name: Convex Agents Usage Tracking
Author: workonthestreets

byworkonthestreets

0•

Analytics & Monitoring

Monitors and logs LLM token consumption and costs within Convex agent applications for billing and optimization.

This skill provides a comprehensive framework for tracking Large Language Model (LLM) usage metrics across Convex-powered agents. It enables developers to log token counts, calculate real-time costs based on provider pricing, and attribute resource consumption to specific users or teams. By implementing structured usage handlers and database schemas, it facilitates automated billing, cost-trend analysis, and proactive alerting to help developers manage their AI budgets effectively.

Key Features

01Real-time usage alerting to identify and prevent budget overruns

02Token-level usage logging for input, output, and total consumption

03Detailed attribution tracking by user, agent, thread, and model

040 GitHub stars

05Provider-specific cost calculation for accurate financial monitoring

06Automated monthly invoice generation via scheduled cron jobs

Use Cases

01Generating detailed cost reports for different departments or client teams

02Implementing a SaaS billing model based on AI token consumption

03Analyzing agent efficiency to optimize prompt engineering and model selection

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add workonthestreets/wolo convex-agents-usage-tracking

For use in Claude.ai and ChatGPT

Key Features

01Real-time usage alerting to identify and prevent budget overruns

02Token-level usage logging for input, output, and total consumption

03Detailed attribution tracking by user, agent, thread, and model

040 GitHub stars

05Provider-specific cost calculation for accurate financial monitoring

06Automated monthly invoice generation via scheduled cron jobs

Use Cases

01Generating detailed cost reports for different departments or client teams

02Implementing a SaaS billing model based on AI token consumption

03Analyzing agent efficiency to optimize prompt engineering and model selection

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add workonthestreets/wolo convex-agents-usage-tracking

For use in Claude.ai and ChatGPT