Mappedinfo/dsh-tool-vision-read1

@deepseek-ai/dsh-tool-vision-read

DSH plugin: vision_read — route image reading to a dedicated vision model (e.g. Kimi K3) so text-only agents can see images

包名
@deepseek-ai/dsh-tool-vision-read
版本
0.1.0-rc.5
许可证
MIT
最近更新
2026年8月16日

安装

此插件尚未提供可验证的 bundle,或兼容性检查未通过。请先阅读仓库说明。 阅读完整 README ↗

Configuration

KeyTypeDefaultMeaning
providerstring— (required)Registered provider route owning the vision model.
modelstring— (required)Vision model id on that route.
toolNamestringvision_readModel-facing tool name.
mode'direct' | 'subagent''direct'Execution mode.
maxImageBytesnumberattachment limitsCap on image bytes sent to the vision route.
maxOutputTokensnumber1024Cap on the vision route's output tokens.
promptstringsee sourceInstruction sent beside the image; {{path}} and {{focus}} placeholders.