dsh-sight
Run the following command in DeepSeek Harness:
dsh plugin install Fu3rte/dsh-sight
Paste the following prompt into your AI chat to install this plugin:
In DeepSeek Harness, you can install this plugin by running the command 'dsh plugin install Fu3rte/dsh-sight', with the source code available at https://github.com/Fu3rte/dsh-sight.
About this plugin
Text-only models in DeepSeek Harness struggle with image understanding. The dsh-sight plugin addresses this by letting users paste images and receive text descriptions via a built-in VLM backend, eliminating the need to switch models and seamlessly enhancing the model's capabilities. Its core strengths lie in flexibility and ease of use: it includes preset VLM backends like OpenCode Zen and Gemini Flash, along with support for custom OpenAI-compatible endpoints; users can configure the backend through a user-friendly web settings page and benefit from multi-image batch processing, handling up to 10 images in one request. Security is also prioritized, with API key protection, local file size limits, and safe remote content handling. dsh-sight is ideal for users who want to add visual understanding to text models without disrupting their existing workflow, whether you're an AI developer, researcher, or casual user working with DeepSeek Harness who needs to process image information.
Screenshots
Use Cases
- Auto-generate text descriptions after pasting images
- Process multiple images in batch for analysis
- Quickly configure vision backend via web settings
Best For
- AI developers using DeepSeek Harness
- Researchers looking to extend model vision capabilities
- Casual users needing image processing functionality
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.