Privacy

The claudeusage-sync CLI reads your local Claude Code logs in ~/.claude/projects/, sums them locally, and uploads only numeric totals:

• API-equivalent cost in USD
• active hours and session counts
• input, output, and cache token totals
• per-model usage mix
• day-level activity used to draw your contribution grid

The CLI is open source (MIT), so you can read every line of what it sends: claudeusage-sync.

• prompts and responses
• your code and file contents
• raw Claude Code JSONL files
• file paths and project names
• individual message and request IDs (only counts are sent)

To merge usage across your devices, the CLI uploads a one-way hash of your machine name (SHA-256 of hostname + OS username). Your raw hostname and username never leave your machine.

Every public response is filtered through a server-side allow-list, so these are never visible on a profile or the leaderboard:

• your email address and sign-in provider
• machine identifiers and hashes
• sync tokens
• exact timestamps — public profiles are day-level only

We store a coarse, nation-level country code (for example, us) derived once at sign-up from your connection’s region. We never store your IP address and never derive a precise location. If your region is unknown, your country is simply left blank.

You sign in with Google or GitHub. Your email and provider stay private. You can choose a public display name or stay anonymous (shown as anonymous-…), and you can switch between the two at any time.

You can delete your account from the dashboard. Deletion removes your public profile and revokes all sync tokens. Aggregated, fully anonymized counts that cannot be tied back to you may be retained.

claudeusage is an independent project and is not affiliated with, endorsed by, or sponsored by Anthropic. “Claude” is used descriptively to refer to Anthropic’s Claude models.