@phamkhachoabk/dsh-local-vision-describe 0.1.0 → 0.1.1

package/README.md

@@ -23,6 +23,19 @@ The first command pulls in Local Vision itself, so it is the only plugin
 command needed on a fresh machine. Restart the harness afterwards: bundle
 layers are composed at boot.
 
+**Upgrading a profile that already has Local Vision** needs one more command,
+because a profile pins its own `^0.2.0` and that range cannot reach the
+`0.3.0` this bundle needs:
+
+```
+dsh plugin --profile web add @phamkhachoabk/dsh-local-vision@latest
+```
+
+Without it the descriptions never appear: the provider loads and the model
+starts, but the consumer row that would ask it is the old one. That is what
+the `@phamkhachoabk/dsh-image-context` peer dependency is for — the package
+manager says so instead of installing a pair that cannot work.
+
 The second is the part this bundle cannot do for you. MLX needs a Python
 environment, and the checkpoint (about 2.5 GB) downloads itself on the plugin's
 first start. Until both exist, the provider reports itself unusable and Local

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@phamkhachoabk/dsh-local-vision-describe",
-  "version": "0.1.0",
+  "version": "0.1.1",
   "description": "Local Vision image descriptions for DeepSeek Harness: a vision-language model on your own machine tells a text-only model what each image shows, beyond the text OCR can read",
   "type": "module",
   "exports": {
@@ -26,6 +26,7 @@
     "@phamkhachoabk/dsh-vision-describe-mlx": "^0.1.0"
   },
   "peerDependencies": {
-    "@deepseek-ai/cordis": "^4.0.2"
+    "@deepseek-ai/cordis": "^4.0.2",
+    "@phamkhachoabk/dsh-image-context": "^0.3.0"
   }
 }