omp usage statusline

Show your provider's coding-plan quota (5-hour / daily / weekly / monthly windows with reset countdowns) directly in the Oh My Pi (omp) status line.

Rendered result (Z.ai GLM coding plan, lite tier):

⏱ lite · 5h 72% (3h 30m) · 7d 54% (2d 10h)

No extension, no polling script — omp ≥ 18 has a built-in usage status-line segment fed by the same provider-usage system as the omp usage command. The catch: the default status-line preset has no slot for it, so you switch to the custom preset and add the segment yourself.

Install

Copy statusLine.yml into ~/.omp/agent/config.yml (merge with any existing statusLine: key), then restart omp.

The segment list is the default preset's segments verbatim plus usage on the right side, so the line looks exactly like before with the quota readout appended.

Notes

• Shows the active provider's usage — switch models/providers and the readout follows (e.g. zai/* models all share one Z.ai coding-plan quota).
• Works for any provider with a usage integration: Z.ai, Anthropic, OpenAI Codex, Google, etc. Run omp usage to see everything omp can see, or omp gallery --surface=segment --plain to preview all status-line segments.
• Colors escalate automatically as a window fills (normal → warning → error).

How it works under the hood (Z.ai example)

omp fetches provider usage itself. For Z.ai's GLM coding plan the endpoint is:

GET https://api.z.ai/api/monitor/usage/quota/limit
Authorization: Bearer <coding-plan API key>

returning data.limits[] entries (number + unit, where unit 3 = hours and 6 = weeks, percentage used, nextResetTime epoch ms) and data.level (plan tier) — the same numbers the z.ai dashboard renders.