Does this work for both Python and JavaScript?

Yes, the skill automatically detects your project environment and configures the appropriate language-specific Sentry integration.

Which AI SDKs are supported by this skill?

The skill supports OpenAI, Anthropic, Vercel AI SDK, LangChain, LangGraph, Google GenAI, and Hugging Face Hub across JS and Python.

What happens if I use an unsupported AI library?

If a supported SDK isn't detected, the skill provides a manual instrumentation guide to help you set up custom Sentry spans for your AI calls.

Can I monitor specific prompt and response data?

Yes, the skill helps you enable Sentry's recordInputs and recordOutputs options to capture full LLM interaction history.

Sentry AI Monitoring Setup

Name: Sentry AI Monitoring Setup
Author: troykelly

bytroykelly

•

Analytics & Monitoring

Configures Sentry AI Agent Monitoring to track LLM calls, token usage, and model performance across major AI providers.

This skill automates the instrumentation of AI-driven applications using Sentry's observability suite to provide deep insights into agent behavior. It follows a detection-first approach to identify installed AI SDKs such as OpenAI, Anthropic, LangChain, or Google GenAI, and then configures the appropriate Sentry integrations to monitor latency, token consumption, and model costs. Whether working in Python or JavaScript, it ensures tracing is correctly enabled and provides clear paths for both automatic and manual instrumentation of complex AI workflows.

Key Features

01Cross-platform support for JavaScript/TypeScript (Node, Next.js) and Python

02Automatic detection of installed AI SDKs including OpenAI, Anthropic, and LangChain

03Validation of Sentry SDK version compatibility and tracing requirements

04Automated configuration of Sentry integrations for input and output tracking

055 GitHub stars

06Guided manual instrumentation for custom AI models and specialized tool calls

Use Cases

01Setting up observability for production AI chatbots to monitor latency and token costs

02Debugging failing LLM calls by capturing prompts and responses in Sentry traces

03Scaling complex AI agent workflows with LangChain or LangGraph using detailed execution spans

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add troykelly/claude-skills sentry-setup-ai-monitoring

For use in Claude.ai and ChatGPT

Download Skill