Models & Providers

What it is — The plugin integrates ChatGPT (Codex), Claude, Grok, and GitHub Copilot subscriptions as LLM providers for DeepSeek Harness via Settings login, without API keys.

Who it is for — If you are using DeepSeek Harness for complex reasoning tasks and have a Claude or ChatGPT subscription, this plugin is suitable. If you use open-source models or non-subscription LLM providers, you can choose other integration methods.

Watch out — Sandbox test passed: it installed successfully in a fresh profile and registered into the harness profile. The plugin requires providing an API key or token to enable features. Claude supports importing credentials from a Claude Code session or using OAuth login.

The verdict — I would install it because it allows direct use of these subscription models in the model picker without manual token management.

What it is — This is a DeepSeek Harness plugin that registers a commandcode provider route with a live model catalog and reasoning-effort support.

Who it is for — When performing command code LLM tasks in DeepSeek Harness web or terminal interface, users who need live model catalog and reasoning-effort support should install this plugin. If your work does not depend on Command Code LLM providers, then this plugin is not applicable.

Watch out — The installation requires configuration of API key in the profile. It passed sandbox testing: installed in a new profile and registered by harness. Static check shows it will execute shell commands.

The verdict — I will install it because it supports multi-account rotation that automatically switches accounts when usage limit is reached.

What it is — Integrates WorkBuddy desktop app models into DeepSeek Harness for direct DSH chat use with no extra setup, supporting parallel groups from both versions.

Who it is for — If you want to call WorkBuddy GLM-5.3 models in DSH chat on DeepSeek Harness and have WorkBuddy desktop app logged in, you can use it directly. Users who only use built-in models or do not need desktop app support do not need to install it.

Watch out — Sandbox test passed with registration in a fresh profile; no obvious issues found. Shell command execution is a static check note only. Reasoning effort detection may consume a small amount of quota.

The verdict — I would install it because it adds WorkBuddy models to DSH chat with zero extra configuration, but only if you already have the WorkBuddy desktop app logged in.

What it is — Quality gate for AI-generated pull requests that automatically scans TODOs, secrets, commit hygiene, and CI status.

Who it is for — If you use AI coding agents such as OpenAI Codex, Claude Code, or Copilot to open pull requests, this provides automated checks for quality issues. Users who prioritize manual review processes can take another angle by relying entirely on their own quality standards for all PRs.

Watch out — Sandbox test shows successful installation as a regular npm dependency, not a profile layer. No other obvious issues identified.

The verdict — I would install it because it posts targeted check-run annotations for matching AI-generated PRs, assuming the workflow already includes the required permissions for statuses and checks.

What it is — Registers Ark Agent Plan, Coding Plan and postpaid model routes into the native DSH model picker.

Who it is for — In scenarios where you manage Agent Plan or Coding Plan using DeepSeek Harness, this plugin is suitable for integrating Ark capabilities. If you mainly use other local agent tools, you can skip installing this plugin.

Watch out — Sandbox testing has not been conducted yet. No obvious issues were found.

The verdict — I would install it because it directly registers Ark plan functions into the DSH model picker.

What it is — Integrates ChatGPT OAuth and OpenAI Codex models into DeepSeek Harness with optional search and image tools.

Who it is for — For users conducting conversations in the DeepSeek Harness Composer who need ChatGPT OAuth to access Codex models, this plugin is suitable. Users who rely on built-in models and global search providers do not need to install this plugin.

Watch out — Sandbox testing passed: the plugin installs successfully in a fresh profile and is registered by Harness. It will execute shell commands. No obvious issues found.

The verdict — I would install it because it adds the Codex model provider via ChatGPT OAuth while Harness retains control over tools, permissions, approval prompts, attachments, session persistence, compaction, and recovery, but only if you need Codex models.

What it is — Integrates account login, API and custom Provider configuration, model switching, image fallback for text-only models, and token/cost attribution into DeepSeek Harness.

Who it is for — When you need to handle OpenAI Codex browser authorization or custom Provider model switching inside DeepSeek Harness, this plugin supplies the required support. If your work relies only on official API key providers and does not require image description capabilities or advanced model switching, you do not need this plugin.

Watch out — Source installation requires manual approval of the build script. Sandbox testing confirmed successful registration in a fresh profile with no obvious issues.

The verdict — I would install it for its token usage tracking and model switching features, but only if my work involves multiple third-party Provider authorizations in DeepSeek Harness.

What it is — DSH plugin integrates ChatGPT OAuth to provide Codex subscription search and backend quota display.

Who it is for — Users who want to display backend quotas and switch subscription search when using Codex models in DSH are suitable to install this plugin. If users who do not wish to retain DSH original sessions, tools and permissions can choose not to install this plugin.

Watch out — Sandbox test passed, installed in fresh profile and registered by harness. No obvious issues found.

The verdict — I would install it because it passed sandbox testing and requires no Codex CLI, but the prerequisite is having a ChatGPT account with Codex access.

What it is — Scientific Figure Library is a local-first MCP App that imports, reviews, and publishes a scientific figure library on disk for exact template reuse in DeepSeek Harness.

Who it is for — When using DeepSeek Harness for scientific document writing, you need to reuse exact templates for scientific figures. If you do not want a figure library shared across multiple AI hosts, it is not suitable.

Watch out — Sandbox testing passed: it installs successfully in a fresh profile and registers with harness. No obvious issues found.

The verdict — I would install it because it supports local disk management for exact scientific figure template reuse in DeepSeek Harness, but only if your work involves scientific document figure reuse.

What it is — Dockyard DSH is a macOS-only native plugin that provides account pools, model directories, and native request interfaces for multiple official AI providers in DeepSeek Harness.

Who it is for — If you need to manage accounts and models across multiple AI providers in DeepSeek Harness using official OAuth, Dockyard DSH is suitable for you. It can scan local login sessions and provide a unified runtime.

If your workflow does not involve calling models from multiple providers or requires only a single provider setup, then you do not need this plugin because it is designed for multi-provider unified management.

Watch out — Installing from source requires the DeepSeek Harness CLI and pnpm, with manual approval for build scripts. The Windows EXE is built but not yet uploaded. The macOS DMG includes a complete runtime but requires configuring OAuth client IDs for some providers.

No obvious runtime issues were found.

The verdict — I would not install it because it is macOS-specific and focuses on multi-provider official integrations, which may not justify the extra setup for users with simpler needs.

What it is — opencode2dsh registers a native LlmAdapter to expose OpenCode Zen free models to DeepSeek Harness without API keys.

Who it is for — If you are performing reasoning tasks in DeepSeek Harness and want selectable thinking levels, this plugin fits. If you use other model providers for your tasks, this plugin is not required.

Watch out — Installation requires Node.js ≥ 20 and outbound HTTPS to opencode.ai. Sandbox testing showed successful registration in a new profile. No obvious issues found.

The verdict — I would install it because it supports selectable thinking levels and a fallback chain for catalog freshness, but only if my tasks require free reasoning models.

What it is — OpenQuantum is an open-source quantum Agent workspace built on DeepSeek Harness, featuring a desktop client, web UI, messaging integration, and Qiskit/MCP tools with scientific validation.

Who it is for — If you're a researcher or developer who needs to analyze quantum circuits, run algorithms like VQE, or query quantum cloud backends using natural language interfaces, installing OpenQuantum provides a unified workspace for these tasks. Users who do not require quantum computing tools or Agent integration for scientific workflows do not need it, as they can stick to standard development environments.

Watch out — Installation requires cloning the repo, running npm ci, and ensuring Node.js 24+. Desktop client may download Electron files on first launch. Sandbox testing passed with successful registration in a new profile. No obvious issues were found in testing.

The verdict — I would not install it because while it integrates quantum tools, the source installation process and dependencies add unnecessary overhead for most users without quantum-specific needs.

What it is — Native Rapid-MLX provider enabling DeepSeek Harness to read model facts from the server instead of settings.yaml.

Who it is for — When using DeepSeek Harness on an M3 Ultra for multi-step tasks, this plugin enables DSH to fetch model facts from the Rapid-MLX server. If you prefer manual configuration of model facts, you do not need this plugin.

Watch out — Requires Node ≥ 22.15 and a running Rapid-MLX server. Verified: installs and activates as a profile layer on M3 Ultra, registers the rapid-mlx route, and serves real queries. Supports plain chat, single tool call, and multi-step bug-fix task. No obvious pitfalls found.

The verdict — I would install it because it allows DSH to use the server's max_model_len for compaction without manual adjustments.

What it is — Registers OpenCode Zen free models as a native LlmAdapter in DeepSeek Harness.

Who it is for — Users who want to use OpenCode Zen free models while doing conversation activities in DeepSeek Harness are suitable for installing this plugin. Users who do not use the web profile of DeepSeek Harness do not need to install this plugin.

Watch out — Sandbox test passed: registers successfully in a fresh profile. No obvious issues found. An upgrade may be required if using an older version of DeepSeek Harness to ensure compatibility with the plugin registration slot.

The verdict — I would install it because it allows using OpenCode Zen models without an API key.

What it is — This plugin integrates Google Antigravity (agy CLI) models into DeepSeek Harness.

Who it is for — If you need to use Google Antigravity models in DeepSeek Harness for chat, this plugin can provide integration. If you use models from other providers, you can choose not to install this plugin.

Watch out — Sandbox testing passed: it is installed in a new profile and registered by harness. No obvious issues found.

The verdict — I will install it because it is an npm package that can be installed in one line, but it requires the Google Antigravity CLI to be installed on the system.

What it is — Integrates ChatGPT subscription access into DeepSeek Harness through OpenAI Codex sign-in flow.

Who it is for — When you want to use Codex models from ChatGPT subscription in DeepSeek Harness, dsh-codex plugin suits you. If you want to use OpenAI Platform API keys directly instead of ChatGPT subscription, you don't need this plugin.

Watch out — Sandbox test passed: installed in new profile and registered into profile. No obvious pitfalls found.

The verdict — I would install it because it allows access to ChatGPT subscription without OpenAI Platform API key.

What it is — This plugin adds Tencent CodeBuddy model support to DeepSeek Harness.

Who it is for — When developing AI agents with DeepSeek Harness that require Tencent CodeBuddy model integration, this plugin enables model registration and quota display. If your agent work focuses on other models or you prefer pure npm installation, you can skip it.

Watch out — It requires subdirectory placement and executes shell commands. Sandbox testing has not been performed. No obvious issues detected.

The verdict — I would install it if I need Tencent CodeBuddy model support in my DeepSeek Harness setup, since it directly provides the model provider integration.

What it is — Hedgehog builds task dependency graphs to guide Claude Code, Cursor and Gemini CLI in ordered coding.

Who it is for — When building full-stack apps with Claude Code and wanting to follow TDD, Hedgehog locks build order via dependency graphs. If your project does not need an opinionated stack, you can skip it.

Watch out — It installs as an ordinary dependency and does not become a profile layer. Sandbox test shows the install command succeeds. No obvious issues found.

The verdict — I would install it because it provides structured paths for full-stack development, but only after activating it per the README.