AI Agent Hub
Back to plugins
🖥️

dsh-chatgpt-subscription

Client Updated 2026.08.26

Run the following command in DeepSeek Harness:

dsh plugin install Aa728848/dsh-chatgpt-subscription

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

Run the installation command in the DeepSeek Harness terminal, or input the complete source repository URL https://github.com/Aa728848/dsh-chatgpt-subscription in the Plugin Market to automatically fetch and activate the extension.

About this plugin

Are you looking to seamlessly access ChatGPT’s premium models within DeepSeek Harness (DSH) but frustrated by strict interface limits or cumbersome API key management? This plugin bridges that gap by proxying your existing ChatGPT subscription through a secure OAuth session, granting DSH direct access to the Gpt-5.x and Codex model families. Enjoy streaming text generation, reasoning summaries, multimodal inputs, and robust tool-calling capabilities without the need for separate API quotas or third-party proxies.

At its core, the plugin prioritizes security and deep integration. It leverages native OS-level credential storage—Windows DPAPI, macOS Keychain, and restricted 0600 files on Linux—to ensure tokens never exist in plaintext outside the host’s secure memory space or application logs. Coupled with automated quota tracking, connection health checks, and a dedicated sub-agent configuration panel, it enables precise model routing, nested delegation control, and efficient multi-agent workflows right out of the box.

Designed specifically for developers, researchers, and AI enthusiasts who already hold an active ChatGPT subscription, this tool is ideal for anyone seeking a safe, cost-effective way to expand DSH’s modeling capabilities. If you want to integrate your existing subscription into a powerful local automation pipeline while maintaining strict security boundaries, this plugin delivers a polished and reliable foundation for your next AI project.

Use Cases

  • Seamlessly integrate your existing ChatGPT subscription into local workflows to cost-effectively access high-tier reasoning and coding models.
  • Configure deeply nested sub-agents locally to centrally manage reasoning depth and context windows across diverse model families.
  • Monitor monthly usage quotas and connection health in real-time while automatically handling token refreshes and upstream rate-limit retries.

Best For

  • Developers and AI researchers who own a ChatGPT subscription and want to expand their modeling capabilities.
  • Power users needing to flexibly switch between multiple models and control agent hierarchies within a local client.
  • Automation workflow builders who prioritize local data privacy and encrypted credential mechanisms.