dsh-llm-agy
Run the following command in DeepSeek Harness:
dsh plugin install flg1217/dsh-llm-agy
Paste the following prompt into your AI chat to install this plugin:
In DeepSeek Harness, run the command 'dsh plugin install flg1217/dsh-llm-agy' to install the plugin, or visit https://github.com/flg1217/dsh-llm-agy for more details.
About this plugin
When working with DeepSeek Harness for complex tasks, users often need to handle multiple subtasks in parallel, perform deep information retrieval, or allow text models to process image inputs, but native dsh may have limitations in these areas. The dsh-llm-agy plugin addresses these pain points by seamlessly integrating Google's Antigravity CLI (powered by Gemini), infusing your workflow with enhanced agent capabilities.
The core abilities of this plugin include: providing independent reasoning subagents, enabling the main agent to delegate tasks like frontend/UI design to Gemini-driven AGY subagents for parallel processing with reusable long-line sessions; enabling deep web search through AGY's search channel, which completes the full research process from retrieval to synthesis rather than just listing links; allowing pure text models (e.g., DeepSeek V4 Flash) to accept image inputs, where pasting an image triggers AGY to read it on-the-fly and convert it to text descriptions transparently; and a connectivity diagnostics panel for checking installation status, testing connections, and copying installation commands. All these features are implemented through dsh's official extension mechanisms without modifying any source code, ensuring smooth upgrades with dsh.
This plugin is particularly suitable for developers, researchers, or content creators who use DeepSeek Harness and want to boost their productivity. If you frequently need to handle multiple AI tasks simultaneously, engage in deep information mining, or process visual content with text models, dsh-llm-agy offers a streamlined, plug-and-play solution that simplifies workflows and lets you focus on core tasks rather than tool configuration.
Use Cases
- When multiple tasks need parallel processing, delegate to independent reasoning subagents.
- For in-depth web research, use AGY to complete full search and synthesis.
- When text models cannot handle images, enable seamless image input support.
Best For
- Developers using DeepSeek Harness who want to boost workflow efficiency.
- Researchers needing in-depth information retrieval and analysis.
- Content creators involved in image processing and content generation.
Related Plugins
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.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.