feat(vom): two channels to deal with canvas

packages/dsh-plugin-browserskill/src/tools.ts

@@ -497,7 +497,8 @@ export function registerTools(deps: ToolDeps): () => void {
       ? "Capture an indented aria-tree snapshot of the session's active tab. Interactive elements " +
         "carry @eN refs for browser_click / browser_fill. Prefer browser_observe for a richer semantic view."
       : "Produce a semantic VOM observation of the session's active tab (roles, states, perception " +
-        "probes) with @eN refs. Visual surface refs are screenshot-only; DOM refs support interaction. " +
+        "probes) with @eN refs. Visual-only surfaces require image understanding and their refs are " +
+        "screenshot-only; keep using any DOM/AX semantics rendered alongside them. " +
         "Read-only: never submits input.";
     register(
       defineTool({
@@ -761,8 +762,11 @@ export function registerTools(deps: ToolDeps): () => void {
       name: "browser_screenshot",
       description:
         "Capture a PNG screenshot of the session's active tab, or crop to a snapshot ref element. " +
-        "This is the supported operation for visual surface refs emitted by browser_observe. Returns " +
-        "the image itself when the deployment supports image input, otherwise a file path.",
+        "This is the supported operation for visual surface refs emitted by browser_observe. Only use " +
+        "it to interpret a visual-only surface when you can inspect image output. If you lack multimodal " +
+        "or image-reading capability, tell the user to switch to a model with image understanding; do " +
+        "not guess content or coordinates. " +
+        "Returns the image itself when the deployment supports image input, otherwise a file path.",
       parameters: {
         session: SESSION_PARAM,
         ref: {