AI Agent Hub
Back to plugins
🖥️

dsh-usage-daily

Client Updated 2026.08.31

Run the following command in DeepSeek Harness:

dsh plugin install HaoR325/dsh-usage-daily

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install HaoR325/dsh-usage-daily in your terminal to install this plugin; source at https://github.com/HaoR325/dsh-usage-daily

About this plugin

Running day-to-day development on DeepSeek Harness often means token consumption and real-world cost are buried inside session logs. dsh-usage-daily solves that with a persistent, semi-transparent card in the bottom-right corner of the web UI. It automatically aggregates the day total input, output, and cached tokens across every session, counts messages and conversation turns, and estimates the cost using DeepSeek default pricing. All data comes from Harness internal session logs, so there is no extra API call and no API key to configure.

Under the hood the plugin refreshes on a 60-second cadence with a server-side cache to avoid hammering the backend, and every pricing constant can be overridden via environment variables to match your actual contract rate. It also exposes a GET /api/dsh-usage-daily/report endpoint that returns the daily aggregated JSON, giving engineers a clean data source for building usage dashboards, cost trackers, or custom integrations. The whole thing is pure Node ESM with zero external dependencies.

If you are running multi-turn sessions in Harness every day and want a quick answer to how much you burned today, or you are assembling a usage-analytics module in your own pipeline, this lightweight plugin is worth a minute to install.

Use Cases

  • Running multi-turn sessions daily and needing a quick glance at token consumption and estimated cost
  • A team reconciling actual AI spend against contract pricing
  • Pulling the daily aggregated JSON feed into a custom usage dashboard

Best For

  • Developers running conversations in DeepSeek Harness day in, day out
  • Engineering teams that need a clear view of token burn and cost
  • Engineers building secondary integrations on top of Harness session data