# dsh-pseudo-vision

> Tools and a request-time image→text bridge that lets text-only DeepSeek Harness models 'see' via OCR, color statistics, pixel scanning, and image metadata. Reifies the pseudo-vision flow surfaced by @YinsenW_ on 2026-08-20.

## Metadata

- Author: [@DDDFXYqiming](https://github.com/DDDFXYqiming)
- Repo: <https://github.com/DDDFXYqiming/dsh-pseudo-vision>
- GitHub: [DDDFXYqiming/dsh-pseudo-vision](https://github.com/DDDFXYqiming/dsh-pseudo-vision)
- Stars: 1
- Language: TypeScript
- License: [MIT](https://spdx.org/licenses/MIT.html)
- Topics: `deepseek-harness`, `dsh-plugin`, `ocr`, `vision`
- Forks: 0
- Open Issues: 0
- Last push: 2026-08-20T09:43:40.000Z
- Added: 2026-08-20T00:00:00.000Z

## Install

```bash
dsh plugin --profile web add github:DDDFXYqiming/dsh-pseudo-vision
```

---

This document is auto-generated by [deepseek-plugin.org](https://deepseek-plugin.org). HTML page: [dsh-pseudo-vision](https://deepseek-plugin.org/plugins/DDDFXYqiming/dsh-pseudo-vision)