AI Agent Hub
Back to plugins
🧰

dsh-quota-meter

Web Tools Updated 2026.08.14

Run the following command in DeepSeek Harness:

dsh plugin install huolianJiao/dsh-quota-meter

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

Run dsh plugin install huolianJiao/dsh-quota-meter in your terminal to install; the source repository is at https://github.com/huolianJiao/dsh-quota-meter.

About this plugin

Working with the DeepSeek API inside DSH leaves you guessing: how much balance is left, how many tokens the current peak / off-peak tier actually buys, and what each turn really cost. Those numbers are buried in console logs and pricing docs, forcing you to leave your workspace to check them. dsh-quota-meter pins that information to the bottom of the DSH sidebar so it is always one glance away.

Collapse the sidebar and you get a status-dotted badge; expand it for a one-line readout of balance, estimated token quota, and the active billing window. The detail panel lays out total, gifted, and recharged balances side by side, then shows a blended estimate keyed to your active model and peak / off-peak tier along with a 0%–100 % cache-hit range where the current window is highlighted. The live-usage section adds cache hit rate, input / output ratio, and a per-turn cost bar chart in which every bar is priced at its own timestamp—hover for the full breakdown.

Pricing tables, peak-hour windows, and fallback ratios are all editable from the Settings page and take effect immediately without a restart. The API key is resolved exclusively inside the local host process and never shipped to the browser; the balance endpoint listens on loopback only. If you work with the DeepSeek API daily in DSH Web and want continuous, at-a-glance visibility into quota and spend, this lightweight plugin is built for you.

Use Cases

  • Checking remaining API balance and available token quota at a glance during daily DSH Web sessions
  • Comparing peak and off-peak pricing to schedule high-cost workloads in the most economical window
  • Reviewing per-turn costs and cache hit rates to refine prompts and calling strategies over time

Best For

  • Developers who rely heavily on the DeepSeek API within DSH Web
  • Technical teams that need precise control over API call costs
  • Individuals tracking cache hits and peak / off-peak pricing to optimize inference budgets