dshbase

Plugin directory / Developer / dsh-openai-oauth

dsh-openai-oauth

Verified · install-tested on dsh DGPisces

✓ Actively maintained Builds on 3 official DSH packages

View on GitHub ↗ ← Back to plugin directory

5Stars
0Forks
0Open issues
JavaScriptLanguage
2026-08-14Last push
Cross-platformPlatform

What it does

DeepSeek Harness provider for GPT models using managed ChatGPT OAuth through Codex app-server

✅
Our take
Works — verified, early-stage project

DeepSeek Harness provider for GPT models using managed ChatGPT OAuth through Codex app-server It installs cleanly and boots without issues in our testing. It's early-stage but functional.

“Verified” means our automated CI actually ran dsh plugin add in a clean profile and it booted — nothing more. Feature descriptions and version compatibility are the author’s claims. This is not a security audit and not an endorsement of third-party code.

README

dsh-openai-oauth

English | 简体中文

Use GPT models from your ChatGPT account as the main model in DeepSeek Harness. Harness keeps its own agent loop and runs its own tools. This package connects its LLM interface to the official local Codex app-server, which handles ChatGPT login, credential storage, token refresh, and model access.

You do not need an OpenAI API key. The package does not copy another Codex installation's auth.json, implement its own OAuth client, or call an unpublished ChatGPT endpoint.

Requirements

  • Node.js 22.19 or newer
  • DeepSeek Harness developer preview 0.1.0-rc.6
  • A ChatGPT account with Codex access

Install

Run one command:

npx -y dsh-openai-oauth install

The installer registers the package with the web and headless profiles, plus any custom profiles that already exist. It also replaces the old deepseek-harness-openai-oauth and dsh-llm-codex-app-server packages if found.

Run the same command to update the package later. If you create a custom profile after installation, run it once more to register that profile.

Sign in with ChatGPT

Start the Harness web interface:

npx @deepseek-ai/dsh web

Open Settings > OpenAI OAuth, select Sign in with ChatGPT, and finish authorization in the OpenAI page. The GPT models available to your account will then appear in the normal Harness model picker.

For a headless profile, sign in from the terminal:

npx @deepseek-ai/dsh plugin --profile headless exec dsh-codex-login

The login is stored separately under ~/.deepseek-harness/codex. Codex owns the credentials and refreshes them when needed.

Select a model

Choose the provider and model in the Harness interface, or set them in ~/.dsh/settings.yaml:

agent-default-model:
  provider: openai-codex
  model: gpt-5.6-sol
  reasoningEffort: high

Run Harness as usual:

npx @deepseek-ai/dsh --profile headless "inspect this repository"

The plugin reads the model list from the signed-in Codex account. It does not keep a hardcoded list of GPT versions.

Uninstall

Remove the package from every current Harness profile while keeping the ChatGPT login for a future reinstall:

npx -y dsh-openai-oauth uninstall

For a clean uninstall that also signs out and deletes ~/.deepseek-harness/codex:

npx -y dsh-openai-oauth uninstall --purge-auth

These commands also remove registrations left by the old package name. They do not install a permanent global npm package. npm may retain its normal download cache.

Local development

npm install
npm test
npm pack --dry-run

This is an independent community plugin. It is not endorsed by DeepSeek or OpenAI.

License

MIT

Install

🧩 Let your agent install it (recommended)

Install the catalog once, then DeepSeek Harness can find and install any plugin from this site automatically:

dsh plugin add dshbase-catalog

Then say "install dsh-openai-oauth for me" — your agent finds it in the directory and installs it. Docs: dshbase-catalog · verified packs.

Web profile:

dsh plugin --profile web add dsh-openai-oauth

Headless (CLI) profile:

dsh plugin --profile headless add dsh-openai-oauth

Package

npm: dsh-openai-oauth · version — · tested on dsh 0.1.0-rc.6

Test report

Verified end-to-end: L1 install + L2 load + L3 runtime Q&A on dsh 0.1.0-rc.6.

When to use it

Extend the agent's coding surface — give it a new tool, workflow, or integration so it handles a dev task it couldn't before.

Who it's for

Developers who want dsh to behave like a teammate on real codebases — editing, running, and verifying changes rather than just answering.

For developers — extending it

The tool/command surface is the seam: expose more of the SDK, add smarter context wiring, or tighten the loop between code changes and verification.

Security: not yet scanned — our daily static scan will cover it shortly.

Share this badge

More in Developer

Browse all 7797 plugins →