modlens
liustack/modlens
Vision bridge for text-only models: paste an image, get structured JSON evidence (OCR, layout, semantics).
DeepSeek Harness plugin
Composer-attached images are transcribed to text by an OpenAI-compatible vision model before reaching text-only DeepSeek models.
Jump to installReview the source first, then copy the community registry command.
dsh plugin --profile web add github:ximengxiaolan/dsh-vision-bridgeThis command comes from the community registry snapshot. Review the repository and package scripts before running it.
Read-only discovery metadata. Verify every field against the repository before use.
registry-record.jsonDiscovery metadata · verify at source{
"name": "dsh-vision-bridge",
"owner": "ximengxiaolan",
"category": "tools",
"description": "Composer-attached images are transcribed to text by an OpenAI-compatible vision model before reaching text-only DeepSeek models.",
"install": "dsh plugin --profile web add github:ximengxiaolan/dsh-vision-bridge",
"npm": null,
"stars": 1,
"added": "2026-08-14",
"source": "registry"
}Catalog inclusion and GitHub stars do not certify this plugin. Inspect its manifest, install scripts, permissions, network behavior, secrets handling, license, and open issues before adding it to a profile.