fix: skip image understanding when primary model has vision

When the primary model supports vision natively (e.g., Claude Opus 4.5),
skip the image understanding call entirely. The image will be injected
directly into the model context instead, saving an API call and avoiding
redundant descriptions.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

src/media-understanding/runner.ts

@@ -4,6 +4,11 @@ import os from "node:os";
 import path from "node:path";
 
 import type { ClawdbotConfig } from "../config/config.js";
+import {
+  findModelInCatalog,
+  loadModelCatalog,
+  modelSupportsVision,
+} from "../agents/model-catalog.js";
 import type { MsgContext } from "../auto-reply/templating.js";
 import { applyTemplate } from "../auto-reply/templating.js";
 import { requireApiKey, resolveApiKeyForProvider } from "../agents/model-auth.js";
@@ -986,6 +991,26 @@ export async function runCapability(params: {
     };
   }
 
+  // Skip image understanding when the primary model supports vision natively.
+  // The image will be injected directly into the model context instead.
+  if (capability === "image" && params.activeModel?.provider) {
+    const catalog = await loadModelCatalog({ config: cfg });
+    const entry = findModelInCatalog(
+      catalog,
+      params.activeModel.provider,
+      params.activeModel.model ?? "",
+    );
+    if (modelSupportsVision(entry)) {
+      if (shouldLogVerbose()) {
+        logVerbose("Skipping image understanding: primary model supports vision natively");
+      }
+      return {
+        outputs: [],
+        decision: { capability, outcome: "skipped", attachments: [] },
+      };
+    }
+  }
+
   const attachmentPolicy = config?.attachments;
   const selected = selectAttachments({
     capability,