FAQ

Is claude-telemetry free?

Yes. All infrastructure runs on free tiers:

• Supabase free tier: 500 MB database, 50k monthly active users, unlimited API requests
• Cloudflare Pages free tier: unlimited sites, 500 builds/month, global CDN
• Cloudflare Workers free tier: 100k requests/day

For personal use, you will never exceed these limits.

Does it work with Claude Desktop?

No. claude-telemetry uses ccusage to extract data from Claude Code's log files. Claude Desktop (the GUI app) uses a different format and storage location. Only Claude Code (the CLI/IDE tool) is supported.

Can I use it with a team?

Currently, claude-telemetry is designed for individual use — one person tracking their own usage across multiple machines.

Team support is planned for v0.3.0, which will include:

• Shared team dashboard with per-member breakdown
• Admin roles and permissions
• Team-level budgets and alerts

How often does the agent sync?

Every 15 minutes by default. You can change this in ~/.claude-tracker/config.json:

{
  "sync_interval_minutes": 5
}

Or run the daemon with a custom interval:

claude-tracker daemon --interval 5

Minimum recommended interval is 5 minutes. More frequent syncing has negligible impact on performance since only new sessions are transmitted.

What data is collected?

Only usage metadata — never prompt or response content:

• Token counts (input, output, cache read, cache write)
• Cost in USD (calculated from token counts and model pricing)
• Model name (e.g., Opus 4, Sonnet 4)
• Session timestamps (start, end, duration)
• Project directory path (which folder you ran Claude Code from)
• Machine identifier (auto-generated UUID + friendly name)

See the Security page for full details.

How are costs calculated?

claude-telemetry uses the same pricing data as ccusage. Token counts are multiplied by the per-model API pricing from Anthropic's published rate card. This gives you the API-equivalent cost of your usage.

If you're on a subscription plan (Pro, Max5, Max20), the actual cost is your monthly subscription fee. The API-equivalent cost tells you how much you'd pay if you were using the API directly — useful for understanding the value of your subscription.

Can I export my data?

Yes. The dashboard supports exporting to:

• CSV — for spreadsheets
• JSON — for programmatic access

You can also query Supabase directly using the REST API or SQL.

What happens if the agent goes offline?

Nothing breaks. The agent is stateless between syncs:

1. If the network is down, the sync fails silently and retries next interval
2. Claude Code keeps logging locally regardless of the agent
3. When the agent comes back online, it picks up from where it left off
4. No data is lost — the local logs are the source of truth

Can I view local data without syncing?

Yes. Use the local command:

claude-tracker local --daily

This shows your usage data from local log files without pushing anything to Supabase.

How do I uninstall?

Step 1 — Remove agent (each PC)

# Stop and remove the service
claude-tracker uninstall-service

# Remove config and data
claude-tracker uninstall

# Or manually:
# Linux/macOS: rm -rf ~/.claude-tracker
# Windows: rd /s /q %USERPROFILE%\.claude-tracker

# Remove the package
pip uninstall claude-usage-tracker

Step 2 — Delete Cloudflare Pages

npx wrangler pages project delete claude-telemetry

Or: Cloudflare Dashboard → Workers & Pages → claude-telemetry → Settings → Delete project.

Step 3 — Delete Supabase project

Supabase Dashboard → Project Settings → General → Delete project. This permanently deletes all data.

After these 3 steps, nothing remains — no data, no services, no secrets.

Where can I get help?

• GitHub Issues — bug reports and feature requests
• GitHub Discussions — questions and community