Usage tracking
What it is
- A local model usage history for calls made by Fased.
- Aggregates chat, channel, task, CLI/system, and worker/subagent runs when usage records exist.
- Groups by provider, model, Agent, channel, task, session, and source.
- Shows input, output, cache read/write, total tokens, and cost when pricing exists.
- Marks usage as unpriced when pricing is unknown instead of hiding the record.
Where it shows up
- Usage page (
/usage): seven-day local usage history with filters for all Agents or a selected Agent. - Dashboard (
/dash): compact seven-day token widget. - Agent Setup: compact Usage card for the selected Agent.
/statusin chats: status card with session tokens and estimated local cost when pricing exists. Provider usage shows for the current model provider when available./usage off|tokens|fullin chats: per-response usage footer (OAuth shows tokens only)./usage costin chats: local cost summary aggregated from Fased session logs.- CLI:
fased status --usageprints a full per-provider breakdown. - macOS menu bar: “Usage” section under Context (only if available).
Providers + credentials
- Anthropic (Claude): OAuth tokens in auth profiles.
- GitHub Copilot: OAuth tokens in auth profiles.
- Gemini CLI: OAuth tokens in auth profiles.
- OpenAI sign-in: OAuth tokens in auth profiles (accountId used when present).
- MiniMax: API key (
MINIMAX_CODE_PLAN_KEYorMINIMAX_API_KEY); coding plan usage uses the provider’s coding-plan window when available. - Xiaomi: API key when configured; provider quota surfaces may be empty.
- z.ai: API key via env/config/auth store.