DSH Vision Toolkit — vision-skills
Give text-only DSH agents visual workflows: image Q&A, OCR, screenshot-to-UI reconstruction, and pixel verification.
Free
Works with the AI Agents you already use
About this skill
DSH Vision Toolkit gives text-only agents in DeepSeek Harness a structured visual workflow, from reading an image to rebuilding and verifying a user interface.
This free package contains the original vision-skills Skill and five workflow references from Anionex/dsh-vision-toolkit v0.1.47, with SKILL.md metadata and an explicit runtime requirement added for this listing.
What it covers:
- Image questions, OCR, target localization, and element detection.
- Long screenshots, scrolling pages, and chat-history extraction.
- Screenshot or sketch to editable UI, with rendered comparison.
- Icon and graphic extraction, transparent PNGs, and SVG tracing.
- Diagrams and whiteboards to structured representations.
- Screenshot-guided GUI work and verification after each action.
- Cropping, dominant colors, HTML screenshots, and pixel diffs.
Runtime requirement: this is the Skill layer of a DSH plugin. Install @anionex/dsh-vision-toolkit in your DeepSeek Harness Profile and configure a vision provider before use. The ZIP does not include the plugin runtime or credentials. Loading this Skill into another agent does not provide the required native vision_* tools.
The Skill is free and MIT-licensed. External vision API usage is billed separately by your chosen provider, if applicable.
Source and installation instructions: https://github.com/Anionex/dsh-vision-toolkit
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 29 days ago
- Free with an Agensi account
Requires
Permissions
Reads user-selected images and writes derived artifacts. Image Q&A, grounding, detection, and OCR may send image bytes to the configured vision service. Cropping, tracing, foreground extraction, color analysis, and pixel diff run locally. HTML screenshot rendering uses a Chrome-family browser. Credentials are managed by the DSH plugin and are not included in this package.