dsh-clinebot

ClinePass is a flat-rate developer subscription that delivers multiples of the standard rate limits for coding and reasoning models through a single OpenAI-compatible endpoint. Native integration into DeepSeek Harness, however, runs into several friction points: the endpoint does not expose /v1/models for dynamic discovery, every model ID must carry a cline-pass/ prefix, rolling five-hour and weekly quotas lack any visual representation, and storing API keys in plain settings introduces a security risk.

dsh-clinebot addresses each of these issues with a purpose-built provider bridge. It opens a dedicated settings page inside DSH where you can one-click sync the models included in your subscription plan directly from the official plan API. A live quota dashboard renders progress bars for both the five-hour rolling window and the weekly window, and fires visual warning banners with a reset countdown when usage crosses the 80 percent and 95 percent thresholds. A session telemetry card tracks request counts, estimated token consumption, latency, and the most recent request timestamp in real time. API keys are pasted in the UI and persisted through the credentials API into a local file rather than left in plain settings. A granular model picker lets you toggle exactly which models appear in the chat selector, and typing /cline in any chat session instantly surfaces quota status, active alerts, and session metrics at a glance.

If you already subscribe to ClinePass and work daily inside DeepSeek Harness across several coding and reasoning models, this plugin removes the manual bookkeeping of model lists and quota monitoring so you always know exactly how much headroom you have left.