Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Claude Code does not document a built-in exact token count for each tool invocation. To investigate high token use, use OpenTelemetry to find token-heavy model requests, then correlate those requests with the tools, agents, skills, or plugins involved. Tool activity and result-size telemetry can point to likely contributors, but they are not a per-tool token bill.
What you can—and cannot—measure
Claude Code’s OpenTelemetry exports include token-usage and cost metrics, as well as model-request and tool activity data. This lets you identify expensive or token-heavy requests and inspect the work around them. Anthropic’s monitoring documentation describes this as tracking usage, costs, and tool activity through OpenTelemetry; it does not promise exact token attribution to every file read, shell command, or other individual tool call.
As an Amazon Associate I earn from qualifying purchases.
The distinction matters because model requests consume tokens, while tool calls are observable activity. Anthropic’s tool-use explanation says requests involving tools can include input and output tokens, tool definitions, tool-use blocks, and tool-result content. A tool’s output may contribute to a later model request, but that request can include other context too, so its full token total cannot be assigned to one tool with certainty.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteUse OpenTelemetry to find likely token drivers
1. Enable telemetry and configure a receiver
Set CLAUDE_CODE_ENABLE_TELEMETRY=1 and configure the metrics and logs exporters described in Anthropic’s monitoring guide. You also need an OpenTelemetry collector or monitoring backend configured to receive and retain the exports. The guide’s quick-start configuration is the appropriate starting point; the exact setup depends on your chosen receiver.
#1 Best Overall
- AI CODING USAGE AT A GLANCE Dedicated WiFi desk display for Claude Code users showing your 5 hour session usage and weekly usage as clean color gauges.
- RUNS ON ITS OWN OVER WIFI No laptop app, no browser tab, and no background daemon needed. Once connected, codeMeter updates automatically from your own Claude account.
- FAST GUIDED SETUP Ships pre flashed and ready to configure. Plug in, join the setup WiFi, enter your home WiFi, paste your Claude Code token, and start tracking.
- COLOR LIMIT WARNINGS Easy visual cues shift from green to amber to red as you approach usage limits, with reset countdowns to help you plan your work.
- MULTIPLE VARIATION OPTIONS Choose from available finish and power options, including USB powered and battery equipped versions depending on the selected model.
2. Find where token use clusters
Start with the claude_code.token.usage metric. Its documented dimensions can help you break usage down by token type, user, team, model, skill.name, plugin.name, or agent.name. For example, compare usage by agent or model to see where a team’s consumption concentrates, then narrow the investigation to the relevant sessions or prompts.
3. Inspect individual model requests
For request-level detail, examine the claude_code.llm_request span and api_request event. These include input, output, cache-read, and cache-creation token counts. Where present, use identifiers such as prompt.id, session.id, and request_id to connect records to the relevant prompt or session.
Rank #2
- AI CODING USAGE AT A GLANCE Dedicated WiFi desk display for Claude Code users showing your 5 hour session usage and weekly usage as clean color gauges.
- RUNS ON ITS OWN OVER WIFI No laptop app, no browser tab, and no background daemon needed. Once connected, codeMeter updates automatically from your own Claude account.
- FAST GUIDED SETUP Ships pre flashed and ready to configure. Plug in, join the setup WiFi, enter your home WiFi, paste your Claude Code token, and start tracking.
- COLOR LIMIT WARNINGS Easy visual cues shift from green to amber to red as you approach usage limits, with reset countdowns to help you plan your work.
- MULTIPLE VARIATION OPTIONS Choose from available finish and power options, including USB powered and battery equipped versions depending on the selected model.
4. Correlate high-token requests with tool activity
Once you identify a high-token request, inspect nearby or linked tool spans and events. They report tool names, execution duration, and result-size fields; tool_use_id can join a tool event to related tool records. Look for repeated activity, large results, errors, or unusually long executions. These clues help identify what may have expanded the context or prompted additional model work, but they do not establish an exact token count for each action.
Free tools Windows power users keep installed
One-click scans. No signup required.
Count cached input as well as ordinary input
The input type in claude_code.token.usage excludes cached input. For a fuller total input figure under the documented OpenTelemetry mapping, add input_tokens, cache_read_tokens, and cache_creation_tokens. Ranking requests on uncached input alone can make a request that reused a large prompt cache look smaller than its total input activity.
Rank #3
- AI CODING USAGE AT A GLANCE Dedicated WiFi desk display for Claude Code users showing your 5 hour session usage and weekly usage as clean color gauges.
- RUNS ON ITS OWN OVER WIFI No laptop app, no browser tab, and no background daemon needed. Once connected, codeMeter updates automatically from your own Claude account.
- FAST GUIDED SETUP Ships pre flashed and ready to configure. Plug in, join the setup WiFi, enter your home WiFi, paste your Claude Code token, and start tracking.
- COLOR LIMIT WARNINGS Easy visual cues shift from green to amber to red as you approach usage limits, with reset countdowns to help you plan your work.
- MULTIPLE VARIATION OPTIONS Choose from available finish and power options, including USB powered and battery equipped versions depending on the selected model.
Choose the right source for the question
| Question | Use | What it establishes |
|---|---|---|
| Which requests or groups are using more tokens? | OpenTelemetry metrics, request spans, and events | Token counts and documented dimensions for analysis; tool activity can be correlated as an investigative clue. |
| Which individual tool call used exactly how many tokens? | No documented built-in per-tool token ledger is established in the current references | Tool names and result-size fields can help investigate likely contributors, not prove exact per-call token attribution. |
| What was officially billed? | Claude Console or the configured API provider’s billing records | Authoritative billing data. Claude Code cost telemetry is an approximation, not the invoice. |
Why the CLI is not a per-action leaderboard
The current Claude Code CLI reference documents print mode (-p), JSON and stream-JSON output, verbose turn-by-turn output, and the print-mode --max-budget-usd limit. It does not document /cost, /stats, or a built-in leaderboard assigning exact token totals to tool calls. Check the installed version’s current help output before relying on commands mentioned in older guides.
JSON or stream-JSON output and the Agent SDK can be useful integration surfaces for scripted workflows. Their existence alone does not establish that a command or SDK output provides a per-tool token ranking; the documented attribution described here comes from the monitoring telemetry.
Quick Recap
Best Value
Rank #4
- Easy Tracking: Easily track your initiative and quickly change your initiative’s order by removing and sliding your markers into place, ensuring smooth and organized gameplay.
- Marker Compatibility: The white acrylic markers can be written on with dry erase markers, allowing for easy and quick changes to the initiative order.
- Flexible Setup: Measuring 11” x 5.25” x 0.75”, this tracker can be used both standing vertically and laying flat, providing flexibility in how you arrange your game space.
- Versatile Usage: Includes three different colored pegs to track rounds and events, enhancing gameplay clarity and organization.
- Made in the USA: Crafted from top-quality durable Birch wood, these stands are designed and manufactured in the USA, ensuring superior craftsmanship.
A practical investigation checklist
- Enable OpenTelemetry and confirm that a configured collector or backend is receiving the data.
- Use
claude_code.token.usageto locate high-use users, teams, models, agents, skills, or plugins. - Inspect request-level input, output, cache-read, and cache-creation counts.
- Include cache-read and cache-creation tokens when comparing total input.
- Correlate high-token requests with tool events and investigate large or repeated results as possible contributors.
- Verify actual charges through the provider’s billing records rather than treating Claude Code cost telemetry as an invoice.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
