ccusage.com

ccusage is a usage analysis tool for coding (agent) CLIs. It reads usage data from the coding CLI tools you already run and turns it into a consolidated view, so you can see how much you're actually using across supported sources instead of checking each tool separately. It's relevant if you use one or more coding agent CLIs and want a clearer picture of your activity and associated costs.

What problem it solves

Coding agent CLIs tend to keep their own usage records in their own formats and locations. If you use more than one, or want to understand cost rather than just raw activity, you end up piecing things together manually. ccusage aggregates that data into a single analysis view.

Core capabilities

Based on the site's own section headings, ccusage organizes its functionality around these areas:

All Sources by Default

The default behavior is aggregation: rather than asking you to pick a source first, ccusage detects supported coding (agent) CLIs and reports their usage together. This is the fastest way to get an overall picture.

Focused Views

When the aggregate is too broad, focused views let you isolate what you care about — for example, a single CLI or a specific subset of usage.

Local Data Sources

ccusage draws on local data sources, meaning the analysis is based on usage records present on your own system rather than a remote account dashboard.

Cost Analysis

Beyond counting usage, ccusage includes cost analysis, which is the piece most relevant if your concern is spend rather than volume.

Enhanced Display

An enhanced display mode presents the results in a more detailed or readable format than a bare summary.

Who it's for

• Developers running one or more coding agent CLIs who want a single consolidated usage view.
• Anyone trying to understand the cost side of their coding CLI usage, not just activity counts.
• Users who prefer analyzing local usage data rather than logging into separate provider dashboards.

What to check before relying on it

• Supported sources: ccusage only aggregates sources it detects as supported. If your CLI isn't among them, it won't appear in the default view.
• Data availability: because it works from local data sources, the analysis is only as complete as the usage records your tools keep locally.
• Cost accuracy: cost analysis depends on the usage data and pricing assumptions behind it; treat the figures as an analysis aid rather than a billing statement.

The site itself is the authoritative reference for the current list of supported sources and the exact commands or views available.

ccusage is a usage analysis tool for coding (agent) CLIs. By default it shows usage from all detected supported sources, so the first thing you see is a combined picture rather than a single tool's numbers. Focused views let you narrow that default view down to a specific source or dimension, and enhanced display changes how the resulting usage data is presented so it is easier to read. Use focused views when the all-sources total is too coarse; use enhanced display when the numbers are correct but hard to scan.

Focused views: narrowing the default all-sources output

The site describes the default behavior as "All Sources by Default" — ccusage detects supported coding (agent) CLIs and aggregates their usage unless you ask for something narrower. Focused views are the mechanism for that narrowing.

In practice, a focused view means you are choosing one slice of the data instead of the whole set. The site groups this under "Focused Views" alongside "Local Data Sources," which indicates the tool reads usage from local data on your machine and that the focused view operates on those local sources.

What a focused view is for

• Isolating one CLI. If you run more than one supported coding agent CLI, the default view mixes them. A focused view lets you look at a single source's usage.
• Isolating one dimension. Rather than a single tool, you may want one aspect of usage across sources — the site lists "Cost Analysis" as a separate heading, so cost is one dimension you can examine on its own.
• Avoiding a misleading total. A combined number can hide the fact that one source dominates. Focusing first, then comparing, gives a more honest read.

How to think about the workflow

1. Start from the default. Run ccusage with no narrowing and note which sources are detected. The default view is your inventory.
2. Pick the slice you need. Decide whether the question is "how much did this one CLI use" or "what did everything cost." Those map to different focused views.
3. Read the focused result. The output now reflects only the selected source or dimension, so totals and per-item figures refer to that slice.

The site does not publish exact flag names or menu paths in the material available, so treat the specific command syntax as something to confirm against the tool's own documentation rather than guessing.

Enhanced display: making the same data readable

Enhanced display does not change what is measured — it changes how the measured usage is shown. The site lists "Enhanced Display" as its own heading, separate from the data-source and cost headings, which supports reading it as a presentation layer over the analysis.

What it improves

• Scanability. Raw usage records are dense. An enhanced view typically groups, aligns, or summarizes so you can find the number you want without reading every row.
• Comparison. When you are looking at a focused view, a clearer display makes differences between periods or sources visible at a glance.
• Cost alongside usage. Because "Cost Analysis" is a distinct capability, an enhanced display is where usage and its cost are most usefully shown together.

How it pairs with focused views

Focused views decide what you see; enhanced display decides how you see it. The useful combination is: focus first to remove irrelevant sources, then read the enhanced output so the remaining data is legible. Doing it in the other order — enhanced display over an all-sources view — still works, but you are styling a larger, noisier set.

Choosing between them

Practical notes and limits

• ccusage analyzes local data sources, so what you can focus on depends on which supported CLIs are installed and have usage data on your machine. An empty focused view usually means no local data for that source, not a broken view.
• The available material does not specify pricing, account requirements, or login. Do not assume the tool is free or requires no account based on this page alone — check the site directly.
• Because exact command syntax is not documented in the source material, verify flag names against ccusage's own help output before scripting anything around it.

If your question is "which sources can ccusage analyze," that is a separate, broader question than how the views work; the focused-view mechanism is what you use once you know the sources exist.

ccusage is a usage analysis tool for coding (agent) CLIs. Its default view is designed to show usage from all detected supported sources at once, rather than requiring you to pick a single tool first. The site also references Local Data Sources, which indicates that ccusage reads usage data already present on your machine instead of pulling it from a remote account dashboard. The exact list of supported CLIs is defined by the site's own documentation, so treat the categories below as the way ccusage organizes analysis, not as a fixed roster.

How ccusage decides what to analyze

The tool's behavior follows a detection-then-display model:

1. Detection — ccusage looks for supported coding (agent) CLI data on the local system.
2. Aggregation — by default, everything detected is combined into one view ("All Sources by Default").
3. Focused views — you can narrow the output to a specific source when you don't want the combined picture.
4. Cost analysis — usage is paired with cost-oriented reporting.
5. Enhanced display — output formatting is treated as a distinct feature area.

The practical consequence: if a CLI isn't detected locally, it won't appear in the default view, even if ccusage conceptually supports that kind of source.

What "sources" means here

How to confirm which sources you actually have

Because support is defined by the site and detection is local, the reliable check is to run ccusage and read what it reports:

• If the default view lists a source, ccusage detected it locally.
• If a tool you use is missing, the likely causes are: no local usage data, data stored in a location ccusage doesn't scan, or the tool not being in the supported set.
• Use focused views to verify a single source is being read correctly before trusting an aggregate number.

Choosing how to use it

• Use the default view when you want a single combined usage and cost picture across every detected coding CLI.
• Use focused views when you need to attribute usage or cost to one specific tool, or when an aggregate looks wrong and you want to isolate the cause.
• Check local data first if results seem incomplete — ccusage can only analyze sources it can detect on the machine.

For the authoritative supported-source list and any setup requirements, consult ccusage's own documentation, since the tool's coverage is defined there rather than by assumption.

ccusage reads usage data from local sources by default and provides a cost analysis view that estimates spending based on that usage. It does not connect to external billing systems or payment platforms, so the numbers it shows are derived from locally stored CLI usage records rather than from invoices or account statements.

Where the data comes from

ccusage is described as a usage analysis tool for coding (agent) CLIs, and its documentation lists "Local Data Sources" as one of its core areas. The default behavior is to detect and view usage from all supported sources on the machine — the heading "All Sources by Default" reflects this.

What this means in practice:

• Usage records are read from where the coding CLI already stores them locally.
• No account login or cloud sync is required to see usage.
• Because the source is local, the analysis reflects what happened on that machine, not usage across other devices or accounts.

If a CLI's data is not present locally, ccusage has nothing to analyze for that source.

What the cost analysis view provides

Cost Analysis is a named view within ccusage. Its purpose is to translate recorded usage into an estimated cost, so you can see how much your coding CLI activity is costing rather than only how many tokens or requests were used.

A few things follow from how the tool is scoped:

• The cost figures are estimates computed from usage data, not billed amounts pulled from a provider.
• ccusage does not appear to integrate with payment platforms or external billing — no pricing links or payment platforms are listed in its site data.
• The exact pricing model used for the estimate is not specified in the available material, so treat the output as a usage-based approximation rather than a guaranteed invoice match.

How this fits with the other views

Cost analysis is one of several ways ccusage presents the same underlying local data:

Because all of these draw on the same local records, switching between them does not change the underlying data — only how it is filtered or displayed.

Practical implications

• Privacy: since analysis runs on local data, usage details stay on your machine.
• Scope: results cover only the sources detected locally; other machines or accounts are not included.
• Accuracy: cost output depends on the usage records and the pricing assumptions built into the tool, so verify against actual billing if the figure matters.
• No billing integration: ccusage is an analysis layer over usage, not a billing or payment tool.

If your goal is to understand roughly what your coding CLI usage costs based on local activity, the Cost Analysis view is the relevant one. If you need exact charges, cross-check with your provider's own billing.

ccusage is a usage analysis tool for coding (agent) CLIs. It is aimed at developers who want to see how much they are using these CLI tools and what that usage costs. If you already run one or more supported coding (agent) CLIs and want a single place to review usage across them, ccusage is likely relevant to you. If you do not use any coding (agent) CLI, there is nothing for it to analyze.

The site is organized around a few clear entry points. Start with the default view, then narrow down based on what you care about.

Start with the default view

The site's default behavior is described as All Sources by Default: it shows usage from all detected supported coding (agent) CLI sources at once. This is the natural first stop, because it tells you what ccusage can see on your machine before you filter anything.

Use this view when you want a broad picture: which sources are detected, and how usage is distributed across them.

Pick a view based on your question

Focused Views and Enhanced Display

Focused Views let you move past the all-sources default and look at a narrower slice of usage. Enhanced Display covers how results are presented rather than what is measured. Together these two sections answer "can I get the specific cut I want, and will it be readable?" — which is usually the deciding factor for whether a usage tool is worth adopting.

Cost Analysis

Cost Analysis is the section to check if spending is your main concern. It is where ccusage turns usage into cost figures. If you only care about raw usage counts, you can skip it; if you need to justify or track spend, start here after the default view.

Local Data Sources

Local Data Sources explains where ccusage gets its input. The name signals that analysis is based on data already present locally rather than a remote account dashboard. Read this section if data provenance matters to you — for example, if you need to know what has to be installed or run before any numbers appear.

A practical path through the site

1. Read the top-level description to confirm ccusage targets coding (agent) CLIs.
2. Check the supported sources list to confirm your CLI is covered.
3. Run or review the default all-sources view to see what is detected.
4. Move to Focused Views to isolate the tool or dimension you care about.
5. If cost matters, go to Cost Analysis; if data origin matters, go to Local Data Sources.

What to verify before committing

• Coverage: confirm your specific coding (agent) CLI appears in the supported sources. Nothing else on the site matters if yours is not listed.
• Data availability: check Local Data Sources to understand what must exist locally for analysis to work.
• Output fit: skim Focused Views and Enhanced Display to confirm the cuts and presentation match how you actually want to read usage.

If all three check out, ccusage fits your workflow. If your CLI is unsupported or you have no local usage data to analyze, it is not the right tool for you.