Skip to content
Back to search

dsh-vision

Near-native image understanding for DeepSeek Harness

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/oil-oil--dsh-vision/badge.svg)](/listings/oil-oil--dsh-vision)

HTML

<a href="/listings/oil-oil--dsh-vision"><img src="/api/listings/oil-oil--dsh-vision/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Use the plugin manager built into DeepSeek Harness:

npx @deepseek-ai/dsh plugin --profile web add github:oil-oil/dsh-vision

Restart Harness, then paste or drag images into the composer as usual. The plugin replaces the official `deepseek-official` adapter while preserving its model catalog, settings, and credentials. It also adds a **Vision Recognition** card to **Settings → Plugins → Plugin configuration**.

> DeepSeek Harness is still in Developer Preview. This release supports `0.1.0-rc.6` and `0.1.0-rc.7`; its settings-card registration satisfies both the legacy list Slot and the current keyed Slot without relying on private runtime inspection.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 95 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <p align="center">
  2. <img src="./assets/readme/hero.svg" alt="dsh-vision: native vision passthrough and a vision bridge for DeepSeek Harness" width="100%">
  3. </p>
  4. <p align="center">
  5. English | <a href="./README.zh.md">中文</a>
  6. </p>
  7. <p align="center">
  8. <a href="https://github.com/oil-oil/dsh-vision/actions"><img alt="CI" src="https://img.shields.io/github/actions/workflow/status/oil-oil/dsh-vision/ci.yml?style=flat-square&label=CI"></a>
  9. <a href="./LICENSE"><img alt="MIT License" src="https://img.shields.io/badge/license-MIT-4D6BFE?style=flat-square"></a>
  10. <img alt="DeepSeek Harness" src="https://img.shields.io/badge/DeepSeek%20Harness-rc.6%20%7C%20rc.7-4D6BFE?style=flat-square">
  11. </p>
  12. `dsh-vision` is a plugin for DeepSeek Harness. Vision-capable models keep receiving images natively. When the selected main model is text-only, the plugin asks a separate vision model to observe the original images, then lets the original DeepSeek model produce the final answer.
  13. ## How it works
  14. | Main model | Image path | Final answer |
  15. | --- | --- | --- |
  16. | Supports images | Original images are sent directly, without preprocessing or OCR | Current model |
  17. | `deepseek-official` or another text-only model | A configured vision model observes the original images; its output is injected as untrusted attachment context | DeepSeek |
  18. | Cloud vision unavailable | Falls back to macOS Vision or Tesseract | DeepSeek |
  19. The plugin does not replace the main model selected in Harness. Multiple image attachments are analyzed together, so comparisons and combined evidence work naturally. The user's task is forwarded unchanged instead of being wrapped in a fixed report template.
  20. ## Install
  21. Use the plugin manager built into DeepSeek Harness:
  22. ```bash
  23. npx @deepseek-ai/dsh plugin --profile web add github:oil-oil/dsh-vision
  24. ```
  25. Restart Harness, then paste or drag images into the composer as usual. The plugin replaces the official `deepseek-official` adapter while preserving its model catalog, settings, and credentials. It also adds a **Vision Recognition** card to **Settings → Plugins → Plugin configuration**.
  26. > DeepSeek Harness is still in Developer Preview. This release supports `0.1.0-rc.6` and `0.1.0-rc.7`; its settings-card registration satisfies both the legacy list Slot and the current keyed Slot without relying on private runtime inspection.
  27. ## Configure Vision Recognition
  28. Open **Settings → Plugins → Plugin configuration → Vision Recognition**. Select ZenMux, Alibaba Cloud Model Studio, TokenDance, or OpenRouter, then enter its API key. The same card lets you change the model ID, API endpoint, and image limit.
  29. The API key is stored through Harness's official credential service. It is write-only in the browser: the plugin can report whether a key exists, but never reads it back into the page, chat, settings document, or session log.
  30. Routing follows the user's choice. A provider selected in Vision Recognition is primary for text-only models. Other enabled Harness vision routes, an existing see configuration, and local OCR are failover only. When the current main model supports images, the original images pass through natively and none of these bridge routes are used.
  31. Choose **Automatic** to skip plugin-managed cloud credentials. The bridge then tries image-capable models already configured in Harness, followed by see-compatible private configuration and local OCR. A Harness custom model must declare `image` as an input modality or it remains a text model.
  32. ## Advanced file configuration
  33. Most setups should use the UI. The equivalent non-secret fields live in the existing `llm-deepseek` section of `$DSH_HOME/settings.yaml`:
  34. ```yaml
  35. llm-deepseek:
  36. visionBackend: zenmux
  37. visionBackendModel: qwen/qwen3.7-plus
  38. visionBackendBaseURL: https://zenmux.ai/api/v1
  39. maxImages: 8
  40. ```
  41. Do not put API keys in this file. Save them in the Vision Recognition card or provide the matching environment variable. Changes apply without a restart.
  42. ## see-skill compatibility
  43. If Harness has no usable vision model, the plugin also reads `~/.config/see/config.env`. It supports ZenMux, Alibaba Cloud Model Studio, OpenRouter, and TokenDance. Environment variables override the private config file.
  44. ```bash
  45. export SEE_PROVIDER=zenmux
  46. export ZENMUX_API_KEY=your-key
  47. ```
  48. `SEE_PROVIDER` selects the primary provider. Other providers with configured keys are failover routes only. If no provider is selected and only one is configured, that provider is used.
  49. When no cloud key is available, or every cloud route fails, the plugin tries local capabilities:
  50. - macOS: built-in Vision OCR, with no extra dependency.
  51. - Linux / Windows: Tesseract with the required language data installed.
  52. Local fallback is primarily OCR and is not equivalent to full multimodal understanding.
  53. ## Security boundary
  54. - Original images are sent only to vision services configured by the user.
  55. - Vision output is marked as untrusted observation data; instructions inside an image receive no system authority.
  56. - Generated vision context affects only the current model request and does not rewrite message history.
  57. - API keys are resolved through Harness credentials or the user's private see config and are never written to this repository.
  58. ## Development
  59. ```bash
  60. pnpm install
  61. pnpm check
  62. ```
  63. The project is available under the MIT License. Cloud routing, joint multi-image analysis, and local fallback behavior are based on the MIT-licensed [oil-oil/see-skill](https://github.com/oil-oil/see-skill). The DeepSeek icon comes from the official [deepseek-ai/deepseek-harness](https://github.com/deepseek-ai/deepseek-harness) repository.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.