AI Agent Hub
Back to plugins
dsh-zhipu-coding-plan preview

dsh-zhipu-coding-plan

Client Updated 2026.08.19

Run the following command in DeepSeek Harness:

dsh plugin install tsja2001/dsh-zhipu-coding-plan

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

Run dsh plugin install tsja2001/dsh-zhipu-coding-plan in DeepSeek Harness to install this plugin, source code at https://github.com/tsja2001/dsh-zhipu-coding-plan

About this plugin

You are halfway through a conversation in DSH Web and suddenly wonder how much of your token window is left or whether your MCP tool quota will hold out this month. Cutting to another tab breaks your flow. dsh-zhipu-coding-plan is a lightweight quota dashboard that answers the question without pulling you out of the coding session.

It embeds the five-hour and weekly token progress bars, a per-tool MCP quota breakdown, and model call volumes for the last 24 hours or 7 days directly beside the chat, trace, and stats tabs in DSH Web. The progress bar shifts to a warning shade at 80 percent and an error shade at 95 percent, with a live countdown to the next reset. Tool quotas are itemized by web search, page fetch, Zread, and other MCP tools. Model usage comes with a time-series bar chart and a per-model share ranking so you can instantly see which model is burning tokens.

If you rely on a personal Zhipu GLM Coding Plan subscription and work inside DSH Web daily, the Quota tab appears right after a restart. Your API key is stored only in the local credential file or an environment variable, never rendered in the UI. All quota queries hit the official Zhipu monitoring endpoint directly—no third-party proxy, no telemetry—and swapping keys requires no restart.

Screenshots

Use Cases

  • Check remaining token window and reset countdown mid-session
  • Break down MCP tool quota usage to identify heavy consumers
  • Compare model call volumes and token shares to optimize routing

Best For

  • Daily coding users on a personal Zhipu GLM Coding Plan subscription
  • Technical teams building with multiple models in DSH Web
  • Indie developers who need precise visibility into API spend and usage