Can I use this with my company's existing observability platform?

Yes, the Enterprise Setup mode allows you to route Claude Code telemetry to any centralized OTLP-compliant endpoint such as Honeycomb, Datadog, or a corporate OTel Collector.

What is the local PoC mode in this skill?

The local PoC mode automatically deploys a full monitoring stack—including Prometheus, Loki, and Grafana—using Docker so you can visualize Claude Code metrics on your own machine instantly.

What specific metrics are collected by this skill?

It tracks session duration, token usage (input, output, and cached), real-time API costs in USD, lines of code modified, and Git actions like commits and PRs.

Is Docker required to use this skill?

Docker is only required if you choose Mode 1 (Local PoC Setup) to host the monitoring stack. Mode 2 (Enterprise) only requires network access to your existing infrastructure.

How do I see my data after the setup is complete?

For local setups, you can access the pre-configured Grafana dashboard at http://localhost:3000. Ensure you have restarted Claude Code after configuration to begin data flow.

Claude Code Monitoring & Analytics

Name: Claude Code Monitoring & Analytics
Author: cskiro

bycskiro

•

分析と監視

Configures comprehensive OpenTelemetry monitoring to track Claude Code usage, token costs, and developer productivity metrics.

This skill automates the deployment and configuration of observability stacks for Claude Code, enabling developers and organizations to gain deep insights into their AI-assisted workflows. It provides a turn-key local Proof of Concept (PoC) environment using Docker, Prometheus, and Grafana for immediate visualization, as well as an enterprise mode for integrating with existing corporate telemetry infrastructure. By capturing granular data on token consumption, API spend, and coding output, it helps users optimize resource allocation and audit AI usage across teams.

主な機能

01Automated local PoC deployment with Docker, Prometheus, Loki, and Grafana

02Enterprise integration for centralized OpenTelemetry (OTLP) endpoints

032 GitHub stars

04Real-time tracking of token usage (input/output/cached) and USD costs

05Pre-configured Grafana dashboards for instant data visualization

06Developer productivity analytics including lines of code and Git activity

ユースケース

01Enterprise teams auditing AI usage across departments for cost management

02Individual developers tracking personal API spending and efficiency

03Troubleshooting connectivity and performance between Claude Code and collectors

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add cskiro/claudex otel-monitoring-setup

For use in Claude.ai and ChatGPT

Download Skill