xby-seg
Run the following command in DeepSeek Harness:
dsh plugin install xby-skill/xby-seg
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install xby-skill/xby-seg in DeepSeek Harness to install this plugin, source code at https://github.com/xby-skill/xby-seg
About this plugin
Instance segmentation goes one step beyond object detection: it does not merely tell you what is in an image, but precisely isolates each object pixel by pixel. xby-seg is a DeepSeek Harness plugin built on the YOLO26 model that performs instance segmentation across all 80 COCO categories, emitting bounding boxes, pixel-level masks, confidence scores, and class labels for every detected instance.
Three input channels keep it flexible: drop in an image URL, a BASE64-encoded string, or a local file path to suit different workflows. A built-in key management command lets you set the API key once in chat and it is persisted automatically, surviving restarts without re-configuration.
Whether you are tackling image analysis, annotation assistance, document processing, or any task that demands pixel-level understanding of visual content, xby-seg plugs instance-segmentation capability into your DeepSeek Harness workflow with minimal friction and no standalone inference server required.
Use Cases
- Pixel-level segmentation of objects in images with mask output
- Batch image processing to retrieve bounding boxes, confidence scores, and labels per instance
- Invoke segmentation directly in chat without deploying a standalone inference service
Best For
- Developers needing pixel-level image understanding
- Practitioners doing annotation assistance or document image analysis
- Users wanting to integrate segmentation into a DSH workflow
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
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.