AI Agent Hub
Back to plugins
🧰

dsh-token-stats

Web Tools Updated 2026.08.17

Run the following command in DeepSeek Harness:

dsh plugin install IZRINO/dsh-token-stats

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

Run dsh plugin install IZRINO/dsh-token-stats in your terminal to install the plugin; the source is available at https://github.com/IZRINO/dsh-token-stats . After installing, restart dsh web and open Settings to see the Token Statistics page.

About this plugin

Running many sessions through DeepSeek Harness quickly raises a practical question: where are the tokens going? dsh-token-stats answers it by adding a dedicated Token Statistics page to the Settings panel, pulling usage data from every session into a single, glanceable view.

The plugin aggregates across four granularities — hour, day, week, and month — and renders a multi-series SVG line chart showing input, cache read, cache write, output, and total tokens side by side. A summary endpoint surfaces call counts, cache hit rate, and active days, while a breakdown view groups metrics by model or provider. Historical sessions are backfilled on first load, all data stays on the local machine, and the HTTP API accepts loopback requests only.

It is built for DSH power users juggling coding, debugging, or batch workloads: if you want to see how much the cache is actually saving, why usage spiked in a given window, or how token spend differs across providers, one click on Settings is all it takes.

Use Cases

  • Compare weekly token spend and cache hit rate across models
  • Drill into a daily usage spike to identify the responsible session or model
  • Evaluate cache read/write ratio to decide whether to tune prompts or switch models

Best For

  • Power users running many DSH sessions daily
  • Team leads or architects reporting model spend to stakeholders
  • Engineers quantifying cache savings and tuning throughput