skills.rest
ankur3-101106
ankur3-101106COMMUNITY
Explore all skills by this author →
Website𝕏@ankur3_101106GitHub

explain-usage

Analyze session transcript files to chart token usage by category.

1|Updated Feb 5, 2026
Search Tags:#token usage#transcript analysis#cost analysis#chart#jsonl#claude session#usage breakdown
One-click install
npx skills add https://github.com/ankur3-101106/dotfiles-hypr --skill explain-usage-ankur3-101106
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: explain-usage
Source: https://github.com/ankur3-101106/dotfiles-hypr/tree/main/Claude/local-agent-mode-sessions/skills-plugin/ab436dd0-752a-4b9b-8503-dc6695d483cb/cedd2988-3bab-48aa-884f-cc7b3bfb8e81/skills/explain-usage
Command: npx skills add https://github.com/ankur3-101106/dotfiles-hypr --skill explain-usage-ankur3-101106

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It answers the question of where a session's tokens went by breaking down usage into understandable groups and presenting the result as a simple chart with a plain-language explanation. ## Core Features & Use Cases - Transcript Analysis: Parses Claude session transcript files (*.jsonl) to measure token consumption across different activity groups. - Weighted Cost Measurement: Applies effective-usage weighting, counting cache reads at 0.1x, cache writes at 2x, and output tokens at 5x the cost of regular input tokens. - Grouped Breakdown: Separates usage into system instructions, Chrome tools, MCP connectors, web research, file operations, subagents, and other activity. - Use Case: After a long working session, ask where your tokens went and receive one chart plus a few short bullet points explaining the biggest consumers in everyday language. ## Quick Start Ask the assistant to explain where this session's tokens went and show a usage breakdown chart.

Frequently Asked Questions about explain-usage

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

FAQPage Schema
How do I see where my Claude session tokens went?▼

Ask for a usage breakdown and the skill parses the session's jsonl transcript files, groups token consumption by activity type, and presents one chart with a short plain-language explanation of the largest consumers.

How is token usage weighted in the breakdown?▼

Usage is measured as effective cost rather than raw counts. Cache reads count at about 0.1x, cache writes at about 2x, and output tokens at about 5x the cost of a regular input token.

What usage categories does the breakdown include?▼

Groups include system instructions, Claude in Chrome tools, MCP connectors grouped by name, web research via WebSearch and WebFetch, file operations, subagents, and everything else. Groups not present in the session are skipped.

Does the usage analysis include subagent token consumption?▼

Yes. Subagent transcripts stored as jsonl files in subfolders of the session folder are counted separately, reporting how many subagents ran and how much each one used.

Can the analysis handle connectors with random ID names?▼

Yes. When a connector's name looks like a random identifier, the skill labels it by what the connector actually does instead of showing the opaque ID.

Related Skills

Explore More Data & Analytics Skills→
vladolaruvladolaru
6

analyzing-cc-sessions

Parse JSONL transcripts to extract metrics and diagnose subagent behavior.

Community
Advanced
Shakudo-ioShakudo-io

context-optimization

Apply compaction, masking, KV-cache reuse, and context partitioning to reduce token usage.

Official
Advanced
JNZaderJNZader

cost-tracking

Track AI token usage and costs across sessions with budget alerts.

Community
Intermediate
centminmodcentminmod
2.6k

session-metrics

Analyze Claude Code session token usage and cost from JSONL conversation logs.

Community
Advanced