Do I need to be a DevOps expert to use this?

No, the skill automates the configuration and startup process. You only need Docker and Docker Compose installed on your local machine to run the setup.

Does this stack support cost tracking?

Yes, it includes specialized dashboards for token usage analysis and cost projections based on your Claude Code interactions.

How do I access the monitoring dashboards?

Once the stack is successfully deployed, you can access the pre-configured Grafana dashboards by visiting http://localhost:3000 in your browser.

Is my Claude Code data sent to the cloud?

No, this setup is entirely local. All logs, metrics, and traces are stored on your local machine within the .observability directory, ensuring data privacy.

What does this skill actually install on my machine?

It deploys a localized Docker-based stack including Grafana for visualization, Loki for logs, Tempo for traces, Prometheus for metrics, and Grafana Alloy as an OTLP collector.

Observability Stack Setup

Name: Observability Stack Setup
Author: adaptationio

byadaptationio

0•

분석 및 모니터링

Deploys a complete LGTM and Grafana Alloy observability stack for monitoring Claude Code performance, costs, and traces.

This skill automates the local deployment of a comprehensive observability infrastructure specifically tailored for Claude Code. By leveraging Docker Compose, it sets up Grafana, Loki, Tempo, Prometheus, and Grafana Alloy to provide deep insights into token usage, tool execution latency, and error rates. It includes pre-configured dashboards that allow developers to visualize session data, track real-time costs, and debug complex tool call traces, making it an essential tool for anyone building or optimizing workflows within the Claude Code ecosystem.

주요 기능

01Pre-built dashboards for cost analysis, tool performance, and error tracking

02Built-in health checks and automated service verification

03Automated Grafana Alloy OTLP collector configuration for telemetry ingestion

04Persistent data storage for historical analysis and session review

050 GitHub stars

06Full LGTM stack deployment including Loki, Grafana, Tempo, and Prometheus

사용 사례

01Monitoring Claude Code token usage and API costs in real-time to manage budgets

02Debugging tool call failures and performance bottlenecks via distributed tracing

03Setting up local development environments with comprehensive logging and metrics

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add adaptationio/skrillz observability-stack-setup

For use in Claude.ai and ChatGPT

Download Skill