This article is derived from the plugin's already-indexed fields (wiki / faq / compatibility_json / readme), not freshly generated by AI. Source field is noted at the end of each section.

Quick start

dsh-vision-toolkit

— source: plugin_wiki.wiki_content

Install & verify

dsh plugin --profile web add @anionex/dsh-vision-toolkit

Run the command above in your DSH Web Profile. Then enable the plugin in the plugin list.

— source: plugins.install

Key points

• Paste and use it immediately. Paste an image in DSH Web and the text-only route switches to its (Vision Toolkit) variant without manual path copying or model changes.
• A seamless image workflow. Native thumbnails, session history, and workspace paths stay intact; Web can preview artifacts and Headless can continue using the same structured results.
• One command to install. The built-in free Gemini 3.7 Flash vision service is ready after installation, with no API key required.
• Built-in free vision. The shared service works immediately after installation with a quota of 300 images per machine per day.
• Vision guided by intent. The agent extracts evidence for the task at hand, such as “Where is the error?” or “Where is the button?”, instead of returning a generic caption.

— source: plugin_wiki.readme_en (fallback readme_raw)

FAQ

Do I need to apply for an API Key for the vision model after installation?

By default, no. The plugin includes a free Gemini 3.7 Flash vision service (vision.anionex.me), ready to use out of the box for pasting images and asking questions; if you need higher quotas or private endpoints, you can replace it with your own OpenAI/Anthropic compatible service in Settings → Vision Tools.

What to do if I paste an image in Web but the model still says it doesn't support image input?

Usually it's due to page caching or the Profile not switching to the variant with (Vision Toolkit) suffix. Restart the Web Profile and refresh the page, confirm the current model has switched to the (Vision Toolkit) variant; you can also put the image in the session workspace and then trigger it through the /vision-tools Skill.

What to do when the free vision service returns 429?

This is shared capacity temporarily exhausted. Wait for the number of seconds specified in Retry-After in the error response and retry; if this happens frequently, it is recommended to replace it with your own endpoint in Settings (Groq or self-built OpenAI compatible gateway both work).

The error says the image is too large or pixel limit exceeded, how to handle it?

The error will clearly indicate whether it's a byte limit (default 4 MiB) or pixel limit (default 20 million). First use vision_crop or external tools to crop/resize the image before calling the vision tool.

vision_html_screenshot can't find Chrome, what to do?

Just install any of Chrome, Chromium, or Edge browsers. Only the HTML screenshot tool is affected; the other 9 vision tools work normally.

Is video, audio, or camera input supported?

Not supported in the current version. This plugin only handles static images (PNG/JPEG/GIF/WebP) and local HTML screenshots, does not automatically click GUI, and does not handle video streams, audio, or camera capture.

Where is the plugin's runtime data stored?

Pasted images are stored in the .dsh-vision-toolkit/ directory under the session workspace; the plugin's isolated Python virtual environment and runtime cache are placed in $DSH_HOME/cache/dsh-vision-toolkit (falls back to ~/.dsh/cache/dsh-vision-toolkit when DSH_HOME is not set).

How to uninstall or temporarily disable this plugin?

The uninstall command is dsh plugin --profile web remove @anionex/dsh-vision-toolkit; if you need to temporarily disable it, you can set disabled: true in the Profile patch and restart the Profile to take effect.

— source: plugin_wiki.faq_json

Compatibility

• DSH: >=0.1.0-rc.6
• Node: ^22.19.0 || >=24.0.0

— source: plugin_wiki.compatibility_json

Pitfalls

Review the upstream repo before installing. This guide is auto-derived from indexed fields and may lag the latest release. If anything contradicts the official docs, treat the upstream source as authoritative.

— source: general rule