Track Claude Code token usage across local JSONL transcripts.
npx skills add https://github.com/ByronWilliamsCPA/.claude --skill usage-report
Please help me install this Agent Skill. Skill: usage-report Source: https://github.com/ByronWilliamsCPA/.claude/tree/main/.claude/skills/usage-report Command: npx skills add https://github.com/ByronWilliamsCPA/.claude --skill usage-report
Claude Code users lack built-in visibility into their token consumption, model-specific spend, and five-hour billing block usage across projects and sessions, making it difficult to control costs, enforce model selection policies, and avoid unexpected usage overruns.
Use the usage-report skill to generate a weekly summary of your Claude Code token spend and top projects by usage volume.
High-intent search queries and answers about installing and using this skill.
Track Claude Code token usage by parsing local JSONL transcripts to generate per-project and per-session spend summaries without relying on external telemetry backends. Reports group token consumption and estimated costs by project, model, or session.
Five-hour billing block monitoring tracks active block consumption and burn rate from local usage data to prevent overages. It integrates with loop circuit breaker checks to safeguard against unexpected usage overruns within each billing window.
Break down token usage by individual subagents using the agents-observe plugin v0.9.11 for subagent-level attribution. This tracks delegation chains and verifies model compliance for cost control and policy enforcement.
Yes, the ccusage CLI tool is required for standard usage reporting and generating daily, weekly, or monthly cost summaries. The skill applies built-in input validation and safety guardrails for command execution.
Enforce model selection policies by generating weekly spend reports grouped by model from local JSONL transcripts. Team leads can identify if expensive Opus models are used for low-complexity read-only tasks, violating internal cost controls.
Claude Code usage reporting works entirely offline by parsing local JSONL transcripts without external telemetry backends. It calculates token spend, billing block consumption, and per-agent attribution locally with built-in safety guardrails.
sachio222Analyze Claude Code JSONL logs for token usage and tool invocations.
butterflyskiesMonitor Claude Code session token usage and calculate API costs.
lgoodcodeBlock unsafe media and text with tiered enforcement and Redis-backed state.
VMRam95Parse local stats cache files to monitor Claude Code OAuth usage and token consumption.
