Install
$ dsh plugin --profile web add @anionex/dsh-vision-toolkitPlugin Overview & Capabilities
AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.
Key Capabilities
- Image Q&A and focused intent inspection (vision_glance)
- Pixel-exact object grounding and bounding box detection (vision_ground, vision_detect)
- Local image cropping and vector tracing to SVG (vision_crop, vision_trace)
- Pixel diff comparison with heatmap rendering for UI verification (vision_pixel_diff)
- Long-screenshot parsing with duplicate overlap detection and OCR (vision_long_screenshot_ocr)
- Foreground extraction and dominant color palette analysis (vision_extract_foreground, vision_dominant_colors)
- Local HTML page rendering and full-page screenshot capture via headless browser (vision_html_screenshot)
- Transparent variant routing for pasted images in DSH Web
- Bundled vision-skills SOP playbooks for automated multi-step visual tasks
Useful For
- Rebuilding front-end UI and web components from screenshots or hand-drawn sketches
- Extracting chat history and text from long scrolling screenshots
- Extracting logos and visual assets into transparent PNGs or editable SVGs
- Performing GUI automation and visual testing via pixel difference auditing
- Pasting images directly into text-only DSH models for immediate contextual Q&A
Who It Fits
- Frontend developers and UI/UX designers automating page reconstruction
- DeepSeek Harness users needing multimodal capabilities with text-only LLMs
- QA engineers conducting visual regression tests on rendered interfaces
- Developers working with screenshots, diagrams, and graphic assets
Documented Limitations
- Requires a separate vision-capable upstream API endpoint (OpenAI Chat Completions, Responses, or Anthropic Messages format)
- Configured shared storage roots are unsupported on Windows due to ACL validation limitations
- HTML screenshot rendering requires a locally installed Chrome, Chromium, or Edge browser
- First run requires internet access to initialize the isolated Python runtime and PyPI wheels if system Python 3.11+ is absent
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 4 runtime, 16 development, 22 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as MIT.
A root package.json was captured and can be inspected by the registry.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- Anionex/dsh-vision-toolkit
- Registry source
- Registry curated
- Registry classification
- Plugin
- Source checked
- 2a1519c · 2026-10-24
This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.
Repository Activity
- GitHub stars
- 906GitHub stars
- Forks
- 46
- Open issues
- 6
- Last commit
- 2026-10-23
- Last release
- 2026-10-18
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Vision routing and pixel-level tools for text-only DeepSeek Harness agents with built-in free fallback.
View plugin →Automates invoice retrieval from IMAP mailboxes, OCR extraction, categorization, and Excel summary report generation.
View plugin →A DeepSeek Harness vision plugin that turns pasted images into structured OCR, layout, and semantic evidence for supported text-only models.
View plugin →DeepSeek Harness integration for ego-lite Chromium browser, providing 33 structured automation tools and a real-time visual watch panel.
View plugin →