AI Agent Hub
Back to plugins
🖥️

dsh-plugin-zquota

Client Updated 2026.08.19

Run the following command in DeepSeek Harness:

dsh plugin install wuyan19/dsh-plugin-zquota

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

Run dsh plugin install wuyan19/dsh-plugin-zquota in your terminal; the source code is hosted at https://github.com/wuyan19/dsh-plugin-zquota and released under the MIT license.

About this plugin

Managing multiple Zhipu GLM Coding Plan accounts—domestic and international—usually means opaque quotas, tedious manual config edits, and a wall of pre-setup before you can send a single message. dsh-plugin-zquota folds all of that into one settings panel inside DeepSeek Harness: every account 5-hour window, weekly, and monthly MCP quota at a glance, colour-coded thresholds (orange at 70%, red at 90%), reset countdowns precise to the minute, and per-tool MCP breakdowns on hover—no more guessing from raw API responses.

Click Set active and the plugin writes the chosen account API key into the endpoint credential reference, then auto-provisions the model route. On a fresh install the first click creates the entire glm-5.3 model entry (1,000,000-token context, 128,000 max output); on an existing setup it only fills what is missing and never overwrites your explicit values. Because llm-pi-ai re-resolves credentials on every model request, the very next message works—no restart, no waiting.

Built for anyone running Zhipu GLM Coding Plan through DeepSeek Harness, especially developers juggling multiple domestic and international accounts who want zero pre-configuration, instant switching, and a clear security boundary: API keys stay in a local 0600 file, the browser never sees key material, and the entire query path runs in-process over Node native fetch.

Use Cases

  • Juggling multiple GLM Coding Plan accounts across CN and intl endpoints and needing real-time quota visibility with reset countdowns
  • Fresh DeepSeek Harness install where you want to call Zhipu GLM immediately without hand-writing provider and model entries
  • Switching between accounts across projects with a single click and zero process restart

Best For

  • Independent developers or small teams subscribed to Zhipu GLM Coding Plan
  • AI engineering teams routing multiple LLM providers through DeepSeek Harness
  • Tech leads who need centralized oversight of multiple API keys and quota budgets