dsh-vision
UnverifiedAdds vision to text-only DeepSeek via OpenAI-compatible VLM
dsh-eyes
UnverifiedAdds image recognition to DSH text models
by qing98351Aug 16, 2026 updated
About this plugin
Provides vision capabilities to DeepSeek Harness by sending pasted, dropped, or imported images to an OpenAI-compatible vision model for text recognition and insertion into conversations. Supports multi-turn review and multiple provider configurations.
- 视觉与多模态
- direct-dsh-signal
Screenshots & demos
Official real work workflow screenshot provided by qing9835/dsh-eyes.
View original media
Official real work workflow screenshot provided by qing9835/dsh-eyes.
View original media
Official real work workflow screenshot provided by qing9835/dsh-eyes.
View original media
Official real work workflow screenshot provided by qing9835/dsh-eyes.
View original mediaInstallation
Before you install
This plugin has not completed plugin-level verification. Review the upstream README, required permissions, dependencies, and data flow first.
Original installation instructions
Taken from the author’s README. Follow the upstream instructions to install.
Upstream README
Loading upstream installation instructions…
Original documentation
Original README
Read the author’s README in place. Documentation and directory verification remain separate facts.
Loading the original README…
Related plugins
dsh-browser
UnverifiedChrome sidebar extension for direct browser control
mimo-vision
UnverifiedVision bridge for image description