skills.rest
Jakeintech
JakeintechCOMMUNITY
Explore all skills by this author →
GitHub

spend

Answers token spend and cost attribution questions from a local metered ledger.

1|Updated Aug 17, 2026
Search Tags:#claude code#usage tracking#ledger#metering#burn rate#cost attribution#token spend
One-click install
npx skills add https://github.com/Jakeintech/waybill --skill spend-jakeintech
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: spend
Source: https://github.com/Jakeintech/waybill/tree/main/skills/spend
Command: npx skills add https://github.com/Jakeintech/waybill --skill spend-jakeintech

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI coding-agent token spend is hard to trace back to the work it actually funded. This Skill answers questions about where Claude Code tokens went, what a story or epic cost, and how burn is pacing, using deterministic numbers from a local metered ledger instead of estimates. ## Core Features & Use Cases - Spend Queries: Break down token usage by story, epic, model, or week, with unattributed spend always shown rather than hidden. - Story Costing: Report total tokens, cache-read share, and tokens-per-point for a specific tracker key such as PLAT-482. - Burn Pacing: Relay pace data comparing spend against linear and work-weighted targets without extrapolation. - Attribution Inbox: Walk through unresolved attribution items one at a time and apply resolutions, pins, or repo defaults that re-meter sessions. - Use Case: Ask "what did PLAT-123 cost" and receive one line with total tokens, cache-read share, and tokens-per-point, all computed by the engine from local transcripts and git history. ## Quick Start Ask where your tokens went this week and the Skill will meter any new sessions, then report spend by account with unattributed share and open spend.

Frequently Asked Questions about spend

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

FAQPage Schema
How do I find out where my Claude Code tokens went?▼

Run the spend query through the waybill CLI to break down token usage by story, epic, model, or week. The numbers come from a local metered ledger built from transcripts and git history, with unattributed spend always shown rather than hidden.

How do I check what a specific ticket or story cost in tokens?▼

Query the story by its tracker key, such as PLAT-482, to get one line with total tokens, cache-read share, and tokens-per-point if the story shipped. USD figures appear only when priced and are labeled with their pricing version.

What is the attribution inbox and how do I resolve it?▼

The attribution inbox holds token spend the resolver could not assign to an account. Query the inbox, review each item's candidates, then apply a resolution with the resolve command, optionally pinning the session or setting a repo default.

Does this token tracking estimate costs or use real data?▼

All numbers are metered deterministically from local transcripts and git history, never estimated. Derived figures like cache savings are explicitly labeled as derived, and unpriced models are disclosed with their coverage percentage.

What happens if a transcript was pruned before metering?▼

The ledger records a meter_gap exception naming the session whose tokens are missing. If Claude Code telemetry was exported to a file, the OTel fallback can recover the session's totals from that export.

Related Skills

Explore More Software Engineering Skills→
abnegateabnegate
2

profile

Generate a developer profile from git commits and Claude Code session history.

Community
Intermediate
GetFresh-VenturesGetFresh-Ventures

perplexity-invoice-pull

Reconcile Perplexity AI invoices with QuickBooks and export CSV/JSON reports.

Official
Advanced
MuvonMuvon
3

ai-cost-finops

Project LLM token spend and apply caching, batching, and model routing levers.

Official
Advanced
mmcmediammcmedia
2

model-usage

Summarize per-model CodexBar costs from local JSON logs or live CLI data.

Community
Intermediate