Claude Code usage dashboard
for your whole team.

How to check your Claude Code usage

Run /usage inside Claude Code. /cost and /stats are aliases that open the same screen, which has three parts:

• Session: tokens per model, API time and lines changed, with a dollar figure Claude Code estimates at list price. On a subscription, that figure is not your bill.
• Plan usage on Pro, Max, Team and Enterprise: how much of your limits is used, and which skills, subagents, plugins and MCP servers used it.
• Stats for the last 24 hours or 7 days (press d or w), read from this machine's history. Usage on other devices or on claude.ai is not in them.

The Claude Code analytics dashboard for Team and Enterprise

Owners of a Team or Enterprise plan open claude.ai/analytics/claude-code. It shows daily active users and sessions, lines of code accepted, the suggestion accept rate, and a top-10 leaderboard you can export for every user as a CSV.

Install Anthropic's GitHub app and it adds the pull requests and lines shipped with Claude Code's help, a public beta that Anthropic calls a deliberate underestimate. For each engineer's tokens and cost, the docs point to OpenTelemetry or the spend report. API customers get a smaller dashboard at platform.claude.com/claude-code, with each user's spend and lines this month.

The Claude Code Analytics API

Claude Console organizations can pull per-user numbers with an Admin API key. One call returns one UTC day: a record per user or API key with sessions, lines added and removed, commits and pull requests by Claude Code, accepted and rejected edits, and tokens and estimated cost per model. Data lags by up to an hour. On Enterprise, claude.ai usage comes from the Claude Enterprise Analytics API instead, with its own key.

curl "https://api.anthropic.com/v1/organizations/usage_report/claude_code?starting_at=2026-10-04&limit=20" \
  -H "anthropic-version: 2023-06-01" \
  -H "x-api-key: $ANTHROPIC_ADMIN_KEY"

Claude Code monitoring with OpenTelemetry (OTel)

For your own dashboards, Claude Code exports OpenTelemetry metrics and events to a collector you run. An admin can set the same variables for every engineer in the env block of the managed settings file.

The metrics include claude_code.session.count, claude_code.token.usage, claude_code.cost.usage, claude_code.lines_of_code.count, claude_code.commit.count and claude_code.pull_request.count. Each carries session.id and, when known, user.email, so you can group by engineer. Anthropic notes that the cost metrics are approximations.

export CLAUDE_CODE_ENABLE_TELEMETRY=1
export OTEL_METRICS_EXPORTER=otlp       # or prometheus, console
export OTEL_LOGS_EXPORTER=otlp          # events: prompts, tool results, API requests
export OTEL_EXPORTER_OTLP_PROTOCOL=grpc
export OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4317
export OTEL_EXPORTER_OTLP_HEADERS="Authorization=Bearer your-token"
claude

Usage trackers: ccusage, /usage and a team view

ccusage, an open-source CLI, reads the local logs of Claude Code, Codex and 16 other coding agents into daily, weekly, monthly and session reports, plus Claude Code's 5-hour blocks. Like /usage, it sees one machine, and Claude Code deletes transcripts after 30 days unless you raise cleanupPeriodDays.

Local counts can disagree. On the Claude Code version we ran, the cache behind /stats counted 1.9 to 2.7 times the tokens of a deduplicated recount of the same transcripts (September 5 to 12, 2026), because each response's usage is logged once per content block. The status line's own cost estimate came within 8% of our recount.

And one laptop isn't the team: in the 30 days to October 3, our heaviest user's laptop opened 596 of the 878 pull requests we merged (68%). The rest came from three other engineers, Codex and other machines.

bunx ccusage claude daily   # Claude Code tokens and cost by day
bunx ccusage blocks         # usage per 5-hour window

How Zest tracks Claude Code usage

Each engineer installs the Zest plugin, which takes about a minute (setup). Three hooks record session start, session end and every tool call, and each session lands on the team dashboard:

Zest needs no collector and doesn't read OpenTelemetry exports, so the two can run side by side.

• Per session: models, input, output and cache tokens, and a cost estimated from list prices. Plans and token rates are on Claude Code pricing.
• Context and 5-hour and weekly limit use at the end of each session, from the status line data.
• Skills, subagents and MCP tools per session, rolled up per engineer in the AI Stack view.
• The branch, commits and pull request a session touched, so the Zest GitHub App can match it to what merged.
• Secrets and personal data are redacted on the engineer's machine before upload; /zest:ignore turns tracking off for a folder.

What Claude Code usage looks like on our own team

Of the 2,620 pull requests merged in our two repos from July 5 to October 3, 2026, 95% carry an AI agent's signature and about 90% carry Claude Code's.

866 pull requests merged in our two repos from August 12 to October 4, 2026 carry a session-analytics block, written by the agent that opened them. The blocks show the default model changing twice in four weeks: Fable 5 and Opus 5 in August, then Fable 5.1 from September 1, then Opus 5.5 from September 22, which wrote every pull request in October. Opus 5.5's merged in a median 25 minutes and Opus 5's in 47, over different weeks and different work.