Can I set budget alerts for my AI agents?

Yes, the skill provides implementation patterns for setting daily, per-call, and per-session budget thresholds to trigger automated alerts when spending limits are reached.

Is it compatible with existing observability tools?

Absolutely. It includes specific integration guides for Langfuse, LangSmith, and OpenTelemetry to ensure your cost data flows seamlessly into your existing monitoring dashboards.

How does this skill calculate LLM costs?

The skill utilizes a built-in pricing table for major models (including Claude 3.5 Sonnet and GPT-4o) to calculate costs based on input, output, and cached tokens per request.

Does it support multi-agent coordination tracking?

Yes, it is designed to track costs at various granularities, allowing you to attribute spending to specific agents, features, or user IDs within a complex multi-agent system.

Token & Cost Tracking

Name: Token & Cost Tracking
Author: nexus-labs-automation

bynexus-labs-automation

0•

分析与监控

Monitors LLM token usage and real-time costs across agents to ensure budget compliance and operational efficiency.

This skill provides a comprehensive framework for instrumenting AI agents with precise token and cost tracking capabilities. It enables developers to implement per-call, per-session, and per-user attribution using up-to-date pricing for major providers like Anthropic and OpenAI. By integrating with observability platforms like Langfuse or OpenTelemetry, it helps organizations prevent 'bill shock' through automated budget alerts, identify optimization opportunities like prompt caching, and gain deep visibility into the financial performance of multi-agent systems.

主要功能

01Granular attribution across features, agents, users, and organizations

02Advanced budget monitoring and automated threshold alerting

030 GitHub stars

04Seamless integration with OpenTelemetry, Langfuse, and LangSmith

05Cache hit tracking and financial impact analysis for prompt optimization

06Real-time LLM cost calculation with multi-provider pricing support

使用场景

01Preventing unexpected budget overruns in production multi-agent systems

02Optimizing prompt engineering costs by identifying high-spend, low-efficiency models

03Analyzing the ROI and unit economics of specific AI-powered features

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nexus-labs-automation/agent-observability token-cost-tracking

For use in Claude.ai and ChatGPT

Download Skill