OpenCode plugin that adds vision support for text-only models — paste images and the plugin analyzes them via a vision CLI tool (e.g., mmx vision describe)
Version 1.0.1, 1 weekly downloads, rank 2218. Judged from its declared dependencies.
| @opencode-ai/plugin | >=1.0.0 |
|---|---|
| @opencode/plugin | - |
[](https://rafal-one.github.io/opencode-v2-ready/p/opencode-minimax-easy-vision-cli.html)
It depends on the V1 plugin API and will not load in OpenCode 2 until the author ports it. See how to port.