skills.rest
yeaboi-ai
yeaboi-aiOFFICIAL
Explore all skills by this author →
WebsiteGitHub

agents-usage

Generates per-model, per-project AI agent cost reports from local Claude Code session logs.

4|Updated Mar 17, 2026
Search Tags:#claude code#ai agents#token usage#cost tracking#spend analysis#usage report#yeaboi
One-click install
npx skills add https://github.com/yeaboi-ai/yeaboi.ai --skill agents-usage-yeaboi-ai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agents-usage
Source: https://github.com/yeaboi-ai/yeaboi.ai/tree/main/claude-plugin/yeaboi/skills/agents-usage
Command: npx skills add https://github.com/yeaboi-ai/yeaboi.ai --skill agents-usage-yeaboi-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Engineers running AI coding agents have no visibility into what those agents cost. This Skill computes token and cost breakdowns from local agent session logs so you can see where agent spend actually goes. ## Core Features & Use Cases - Cost breakdowns: Aggregates token usage and estimated cost per model, per project, and per source over a configurable window (1-365 days), plus a daily trend. - Billing-aware reporting: Distinguishes subscription plans (API-equivalent estimates) from API billing, and surfaces cache-cost share and fallback-priced shares. - History comparison: Compares current usage against previously generated reports instead of recomputing. - Use Case: Ask what your Claude Code agents cost last month and get a conversational report with totals, top models and projects, and a daily trend — with privacy preserved since only aggregates are read, never session transcripts. ## Quick Start Ask the assistant to show what your AI coding agents cost over the last 30 days, broken down by model and project.

Frequently Asked Questions about agents-usage

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

FAQPage Schema
How do I see what my Claude Code agents cost?▼

Run the agents_usage report, which reads local Claude Code session logs and computes token and cost estimates per model, project, and source. By default it covers the last 30 days; narrow it with window_days, project, or source filters.

How are AI agent usage costs calculated from session logs?▼

Costs are estimated by aggregating token counts from local session logs and pricing them against public rate tables. Tokens are attributed to the day they were spent, producing a real per-day trend, and cache traffic is reported separately as cache_cost_share.

Are the agent cost numbers exact billing amounts?▼

No. The feature is in beta and produces estimates from local logs and public rate tables, not the provider's invoice. On subscription plans the total is an API-equivalent figure included in the plan, and unknown models are priced at a fallback tier.

Does agent usage reporting upload my session transcripts?▼

No. Only aggregate numbers are read into the report; session transcripts are never copied or uploaded. Exports are saved locally under ~/.yeaboi/exports/agentwatch/usage/.

Why does my agent usage report show $0.00?▼

A $0.00 report with a "no local agent sessions" warning means the machine has no Claude Code history to read, not that agents are free. Check the warnings field to understand what informed the output.

Related Skills

Explore More Software Engineering Skills→
yulonglinyulonglin
11

insights

Generate coaching-style HTML usage insights reports from Claude Code session histories.

Community
Intermediate
violetiovioletio

finance-engineer

Analyze infrastructure costs and generate monthly/annual projections with anomaly alerts.

Official
Advanced
mahg-esmahg-es
3

daily-standup

Collect agent updates into a Markdown-formatted daily standup summary.

Official
Intermediate
wyre-technologywyre-technology
39

Pax8 Invoices & Billing

Retrieve Pax8 invoice details and usage summaries for billing reconciliation.

Official
Intermediate