Skill Usage Tracker FAQs

Question 1

How does this skill help monitor and reduce API costs?

Accepted Answer

It tracks four key token metrics: input, output, cache read, and cache creation. By using the built-in analysis scripts, you can identify high-cost tasks and see how effectively Claude's prompt caching is reducing your expenses.

Question 2

What does the Skill Usage Tracker do?

Accepted Answer

The Skill Usage Tracker automates the setup of a local monitoring system for Claude Code. It installs hooks that capture every skill invocation, recording execution duration, the specific user prompts used, and detailed token usage metrics including cache hits.

Question 3

When is the best time to use this skill?

Accepted Answer

You should use this skill when starting a new project or whenever you want to analyze your development patterns. It is particularly useful for identifying repetitive manual tasks that could be automated into new custom skills.

Question 4

Is my activity data kept private?

Accepted Answer

Yes. This skill is designed with a privacy-first approach. All logs are stored locally on your machine in the `.claude/activity-logs/` directory and are automatically added to your .gitignore to prevent them from being uploaded to version control.

Question 5

Does it provide visual reports or just raw data?

Accepted Answer

The skill includes both Python and Bash analysis scripts that process raw JSONL logs into readable reports. These reports summarize your most-used skills, average execution times, and total time saved through automation.

Skill Usage Tracker

Skill Usage Tracker

Key Features

Use Cases

Key Features

Use Cases