Works with the AI Agents you already use
claude code usage dashboard
Analyze local Claude Code transcripts to report token usage and estimate API-equivalent costs.
$5
claude code usage dashboard
Example session with this skill installed
Show me a breakdown of my Claude Code usage and estimated costs for the last 7 days.
- Read your context and instructions
- Compiled the claude code usage
I have analyzed your local Claude Code transcripts from the last 7 days.
Claude Code Usage Report (Last 7 Days)
Window Totals
- Today: 45,201 tokens, $0.68, 12 responses
- Last 7 days: 284,150 tokens, $4.26, 82 responses
- Last 30 days: 1,120,400 tokens, $16.81, 310 responses
By Model
| Model | Input | Cache Write | Cache Read | Output | Cost |
|---|---|---|---|---|---|
| claude-3-7-sonnet-20250219 | 150,200 | 45,000 | 80,000 | 8,950 | $3.45 |
| claude-3-5-haiku-20241022 | 42,000 | 0 | 0 | 5,200 | $0.81 |
Top 5 Projects
- api-gateway: 92,100 tokens ($1.38)
- frontend-ui: 75,400 tokens ($1.13)
- auth-service: 55,200 tokens ($0.83)
- utils-lib: 40,300 tokens ($0.60)
- docs-site: 21,150 tokens ($0.32)
Cost is an API-equivalent estimate at list prices, not your bill. Pro, Max and Team plans are billed differently.
Next steps
- Run
node scripts/usage.mjs serveto view these trends in a browser dashboard. - To see usage for a specific project only, use the
--include <alias>filter. - If you work on multiple machines, use
exportto sync this data to a shared volume.
Connects securely to your tools. The creator never sees your data.
About this skill
Claude Code logs usage data locally, making it difficult to track total spend or compare activity across multiple development machines. This skill parses local transcripts to provide a clear view of token consumption and estimated API costs without sending your data to external servers.
What it does
- Token tracking aggregates input, output, and cache hit/miss tokens from local session logs.
- Cost estimation calculates API-equivalent spending based on current Anthropic list prices.
- Multi-machine sync combines usage from different workstations via a shared folder export.
- Visual dashboard launches a local browser interface for charts and project-level breakdowns.
- Granular filtering isolates usage by specific projects, time ranges, or model types.
How it works
- Verify environment by checking for Node 18+ on the host system.
- Scan transcripts located in the default Claude configuration directories.
- Generate report using the
previewcommand for terminal summaries orservefor the web UI. - Aggregate shards if the user provides a shared directory containing exports from other machines.
Frameworks & tools
Built with Node.js. Uses Chart.js via CDN for the web dashboard. Requires no external database or account.
Why this beats prompting it yourself
Manual calculation of cache-aware token pricing is error-prone and tedious. This skill automates the parsing of complex local session files and handles cross-machine data deduplication that a basic prompt cannot access or process reliably.
Use cases
- Audit weekly Claude Code spend across a distributed engineering team.
- Identify which local projects are consuming the most tokens.
- Compare the cost-efficiency of different models during development tasks.
- Export usage history before decommissioning a temporary dev environment.
Known limitations
Requires Node 18 or higher. Only tracks usage recorded in local transcript files. Estimates reflect list API prices which may differ from Pro or Team plan billing.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.
Requires Node 18+. Pricing estimates use API list rates and may not match Pro/Team plan billing. Shared dashboard requires internet for Chart.js CDN.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 5 days ago
- Software
- cc-logbook
Node.js
Permissions
Allowed Hosts