local-claude-telemetry

Analyze local Claude Code session transcripts, usage metrics, and events.

Updated Feb 10, 2026
One-click install
npx skills add https://github.com/ammonfife/sakima.co --skill local-claude-telemetry
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: local-claude-telemetry
Source: https://github.com/ammonfife/sakima.co/tree/main/.claude/skills/local-claude-telemetry
Command: npx skills add https://github.com/ammonfife/sakima.co --skill local-claude-telemetry

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a comprehensive analysis of local Claude Code data streams, enabling token/cost/usage analytics, session forensics, per-model consumption, and tool-usage patterns, ensuring the health of the local collector.

Core Features & Use Cases

  • Analytics: Read and analyze local-only Claude Code data streams.
  • Session Forensics: Gain insights into sessions, including prompts and responses.
  • Per-Model Consumption: Track usage metrics by model and type.
  • Tool-Usage Patterns: Understand tool-usage patterns and verify collector health.
  • Use Case: For debugging and optimizing Claude Code usage on a local machine.

Quick Start

Analyze the Claude Code data streams on this machine using the local-claude-telemetry skill.

Frequently Asked Questions about local-claude-telemetry

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze local Claude Code session transcripts for debugging?

To analyze local Claude Code session transcripts for debugging, you can use a local telemetry skill to read local-only data streams, enabling session forensics to gain insights into prompts, responses, and tool-usage patterns.

What is the best way to track per-model token usage and cost analytics locally?

The best way to track per-model token usage and cost analytics locally is by analyzing local Claude Code data streams, which allows you to monitor consumption metrics by model and type without sending data to external services.

Can I monitor Claude Code tool-usage patterns without sending data to external services?

Yes, you can monitor Claude Code tool-usage patterns without external services by using a local telemetry skill that operates strictly on local-only data streams to analyze events and verify collector health.

Does local Claude Code telemetry require any external dependencies or API connections?

Local Claude Code telemetry does not require external dependencies or API connections, as the skill operates entirely on local-only data streams to ensure that no session or usage data is sent to external services.

How do I verify the health of my local Claude Code data collector?

To verify the health of your local Claude Code data collector, you analyze local data streams and events to understand tool-usage patterns, ensuring the collector is accurately capturing session transcripts and usage metrics.