dsh-provider-copilot
Run the following command in DeepSeek Harness:
dsh plugin install youngrock-labs/dsh-provider-copilot
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install youngrock-labs/dsh-provider-copilot inside DeepSeek Harness to install the plugin; the full source repository is at https://github.com/youngrock-labs/dsh-provider-copilot .
About this plugin
If you already subscribe to GitHub Copilot but want to unify model calls inside DeepSeek Harness (dsh), dsh-provider-copilot is that bridge. It wraps Copilot's Chat Completions API as a native dsh provider, requiring zero source changes to dsh itself. Once installed, the Copilot model group shows up in the model picker immediately.
Core capabilities focus on day-to-day inference: streaming completions with token-by-token output; model lists computed as the intersection of a local whitelist and the upstream remote catalog; reasoning deltas transparently mapped into dsh's collapsible thinking row. Authentication supports a three-tier fallback—Device Flow login, environment-variable injection, and on-disk OAuth cache—while requesting only the read:user scope.
It is built for individual developers or subscribers who already use GitHub Copilot daily and want to call models directly inside dsh sessions instead of switching to VSCode or a browser. The plugin explicitly excludes tool and function calling, keeping its scope clean: an observable, plain chat-completion inference channel.
Screenshots
Use Cases
- Call GitHub Copilot subscription models directly inside dsh sessions without switching IDEs or browsers
- Inspect model reasoning step by step via reasoning-delta passthrough in dsh collapsible thinking rows
- Manage credentials via COPILOT_TOKEN or /copilot login to switch models with zero extra configuration
Best For
- Individual developers with an active GitHub Copilot subscription
- Small teams that want to consolidate multi-model access into a single dsh workflow
- Inference-focused developers who need streaming output and reasoning visualization without tool calling
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.