AI Agent Hub
Back to plugins
🤖

dsh-openai-oauth

Model Inference Updated 2026.08.21

Run the following command in DeepSeek Harness:

dsh plugin install DGPisces/dsh-openai-oauth

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

In DeepSeek Harness, you can install this plugin by running 'dsh plugin install DGPisces/dsh-openai-oauth', with the full source code available at https://github.com/DGPisces/dsh-openai-oauth.

About this plugin

Integrating GPT models into DeepSeek Harness for inference often presents hurdles like API key dependencies, added expenses, and setup complexities. Users with ChatGPT accounts desire to leverage their subscriptions directly without juggling additional credentials or hardcoded configurations. The dsh-openai-oauth plugin addresses this by using a local Codex app-server to handle OAuth authentication, managing login, credential storage, and token refresh seamlessly. Its core capability lies in the keyless approach: it dynamically fetches available models from the signed-in ChatGPT account, eliminating the need for static lists, while preserving Harness's native agent loop and tool execution. This plugin is perfect for developers working with DeepSeek Harness who have a ChatGPT subscription and aim to streamline model inference, as well as AI enthusiasts or researchers needing to test GPT models locally in a frictionless manner.

Use Cases

  • Use GPT models from ChatGPT directly in DeepSeek Harness for inference.
  • Integrate and test GPT functionalities without needing an OpenAI API key.
  • Rapidly deploy and validate AI models in local development environments.

Best For

  • Developers with ChatGPT accounts seeking to streamline GPT model usage.
  • AI researchers needing to test and experiment with GPT models locally.
  • AI enthusiasts who want to use GPT models without extra configurations.