v0.1.2

npm page now shows the English README only; Chinese docs at docs/README.zh.md.

v0.1.1

English README (README.md) + Chinese README (README.zh.md). Emphasizes: solves the problem of DSH forbidding image uploads to non-multimodal (text-only) models.

v0.1.0

Initial release: image_to_text model tool with configurable vision API providers (custom apiKey/endpoint via plugin config or IMAGE_TO_TEXT_API_KEY_/IMAGE_TO_TEXT_ENDPOINT_ env vars).