Works with the AI Agents you already use
Token Meter
Claude Code tells you very little about how many tokens you actually burn.
$5
Token Meter
Example session with this skill installed
Show me my token usage report for this month and save it to a file.
- Read your context and instructions
- Compiled the token meter
Here is your token report for this month, read from your local Claude Code logs:
Month total: 9.5M tokens over 12 active days (179 model calls)
Busiest day: the 10th, 1.5M tokens
Cache hit rate: 94% of input came from cache
| Day | Total | Output | Cache read |
|---|---|---|---|
| 2026-11-09 | 912,300 | 21,400 | 868,000 |
| 2026-11-10 ★ | 1,507,761 | 33,100 | 1,430,000 |
| Model | Total | Cache hit |
|---|---|---|
| claude-sonnet-5-5 | 5.1M | 95% |
| claude-opus-5-5 | 3.3M | 93% |
Saved to ~/.claude/token-meter/reports/token-report-2026-11.md
(Figures above are an example.)
Connects securely to your tools. The creator never sees your data.
About this skill
Claude Code tells you very little about how many tokens you actually burn. Token Meter reads the session logs Claude Code already keeps on your computer and turns them into answers.
Ask, and get a report
- "Show my token report": one row per day with input, output, cache read and cache write, a per-model table, the month total, the busiest day and your cache hit rate
- "Token report for last month, save it": keeps a Markdown copy, because Claude Code deletes logs after 30 days
- "Make it a card": a dark one-page HTML summary with a calendar heatmap, ready to screenshot
- "How many tokens did I use this week?": a quick plain-language summary
Optional live panel for Claude Code
- A counter above the message box and in the status bar: this session, output, cache hit rate, today
- A Usage panel beside the chat: 7 days / 30 days / All time, a 20-week heatmap, tiles for output, sessions, messages and cache reads, a daily tokens chart with your busiest day, and a per-model split
/token-usageopens it,/token-meterhides the counter. Ask Claude to install it, or run two commands; uninstall the same way
Private by design
- Read-only: only reads the
~/.claude/projectslogs - Offline: no network calls, no telemetry, no API key
- Plain, commented JavaScript you can read before you run it
Requirements and limits (please read)
- Needs Node.js 18+ (Windows, macOS, Linux)
- Counts Claude Code sessions on this computer only. Chats in the Claude app or on claude.ai are not stored locally, so no local tool can count them
- The live panel needs a recent Claude Code version with plugin panel support. The full panel shows in the Claude desktop app; the terminal shows the counter and a text heatmap
- Token totals include cache reads, which cost far less than fresh input; the report shows them separately
Only counts local Claude Code sessions. Logs older than 30 days are lost unless saved manually. Live panel requires plugin-capable Claude Code versions.
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 any payment.
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 3 days ago
Requires