dsh-plugin-vision
Run the following command in DeepSeek Harness:
dsh plugin install tdf1995/dsh-plugin-vision
Paste the following prompt into your AI chat to install this plugin:
You can install this plugin by running the command 'dsh plugin install tdf1995/dsh-plugin-vision', or obtain it from the GitHub source at https://github.com/tdf1995/dsh-plugin-vision.
About this plugin
In AI conversations, many powerful text-based large language models like DeepSeek lack support for image input, making tasks such as image analysis, OCR, and visual question answering cumbersome, often requiring users to switch models or rely on external tools. The dsh-plugin-vision plugin seamlessly adds visual capabilities to text-only models within DeepSeek Harness by integrating free Gemini and GLM vision APIs, introducing a tool called see_image that enables direct analysis of local images in conversations, with features like custom questioning, automatic image compression, and dual-provider failover. It also includes built-in browser paste and drag-and-drop support for images, making visual processing more intuitive and convenient.
This plugin is ideal for DSH users, including developers, researchers, and anyone needing to handle image tasks in text model dialogs. Whether for quick screenshot interpretation, document OCR, or chart analysis, it offers a low-cost or even free solution without the need to frequently switch tools or models, enhancing both efficiency and interaction experience.
Use Cases
- Analyze text content in screenshots
- Perform OCR on documents
- Answer questions about image content
Best For
- DSH users who need to handle image tasks
- Developers testing visual features
- Researchers conducting multimodal experiments
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.