modlensA83

dsh plugin --profile web add @liustack/modlens

The name after --profile is a profile name you choose yourself — swap in whichever one you use.

The flagship DeepSeek and GLM chat models are text-only and cannot read images. ModLens is a bolt-on vision engine that restores that capability with a very short interaction path: paste the image directly into the composer and the plugin takes over through a native modlens_read_image tool. It also auto-discovers every provider route carrying a text-only DeepSeek or GLM model and adds a companion entry suffixed (modlens vision) for each, while automatically excluding both families' own vision models so nothing is taken over needlessly. The author's stated design goal is the lightest possible touch: no hooks, no wrappers, no local proxy daemon, and not a single line changed in any host config — on DSH it is exactly one plugin, and uninstalling is deleting it.

• ▸Two paste routes: plain paste (temp file + modlens_read_image), or pick a (modlens vision) model entry to keep the thumbnail visible
• ▸Auto-generates a vision entry per text-only DeepSeek/GLM route; native vision models are skipped
• ▸No hooks, wrappers or local proxy, and no host config changes; the same author's ModSearch covers the web

npm 已发 45 版、故障排查/输出契约/安全文档齐备，DSH 首个视觉插件，成熟度极高

Drop it into your README — it links back to this page.

Badge (inline)Copy Markdown

[![dshfind](https://dshfind.com/api/badge/liustack/modlens)](https://dshfind.com/en/plugins/liustack/modlens?ref=badge)

Card (440×200)Copy Markdown

[![dshfind](https://dshfind.com/api/card/liustack/modlens)](https://dshfind.com/en/plugins/liustack/modlens?ref=badge)