refactor: normalize media attachment selection

src/media-understanding/apply.test.ts

@@ -424,4 +424,64 @@ describe("applyMediaUnderstanding", () => {
       ["[Audio 1/2]\nTranscript:\nnote-a.ogg", "[Audio 2/2]\nTranscript:\nnote-b.ogg"].join("\n\n"),
     );
   });
+
+  it("orders mixed media outputs as image, audio, video", async () => {
+    const { applyMediaUnderstanding } = await loadApply();
+    const dir = await fs.mkdtemp(path.join(os.tmpdir(), "clawdbot-media-"));
+    const imagePath = path.join(dir, "photo.jpg");
+    const audioPath = path.join(dir, "note.ogg");
+    const videoPath = path.join(dir, "clip.mp4");
+    await fs.writeFile(imagePath, "image-bytes");
+    await fs.writeFile(audioPath, "audio-bytes");
+    await fs.writeFile(videoPath, "video-bytes");
+
+    const ctx: MsgContext = {
+      Body: "<media:mixed>",
+      MediaPaths: [imagePath, audioPath, videoPath],
+      MediaTypes: ["image/jpeg", "audio/ogg", "video/mp4"],
+    };
+    const cfg: ClawdbotConfig = {
+      tools: {
+        media: {
+          image: { enabled: true, models: [{ provider: "openai", model: "gpt-5.2" }] },
+          audio: { enabled: true, models: [{ provider: "groq" }] },
+          video: { enabled: true, models: [{ provider: "google", model: "gemini-3" }] },
+        },
+      },
+    };
+
+    const result = await applyMediaUnderstanding({
+      ctx,
+      cfg,
+      agentDir: dir,
+      providers: {
+        openai: {
+          id: "openai",
+          describeImage: async () => ({ text: "image ok" }),
+        },
+        groq: {
+          id: "groq",
+          transcribeAudio: async () => ({ text: "audio ok" }),
+        },
+        google: {
+          id: "google",
+          describeVideo: async () => ({ text: "video ok" }),
+        },
+      },
+    });
+
+    expect(result.appliedImage).toBe(true);
+    expect(result.appliedAudio).toBe(true);
+    expect(result.appliedVideo).toBe(true);
+    expect(ctx.Body).toBe(
+      [
+        "[Image]\nDescription:\nimage ok",
+        "[Audio]\nTranscript:\naudio ok",
+        "[Video]\nDescription:\nvideo ok",
+      ].join("\n\n"),
+    );
+    expect(ctx.Transcript).toBe("audio ok");
+    expect(ctx.CommandBody).toBe("audio ok");
+    expect(ctx.BodyForCommands).toBe("audio ok");
+  });
 });

src/media-understanding/attachments.ts

@@ -7,7 +7,7 @@ import { fileURLToPath } from "node:url";
 import type { MsgContext } from "../auto-reply/templating.js";
 import type { MediaUnderstandingAttachmentsConfig } from "../config/types.tools.js";
 import { fetchRemoteMedia, MediaFetchError } from "../media/fetch.js";
-import { detectMime, getFileExtension, isAudioFileName } from "../media/mime.js";
+import { detectMime, getFileExtension, isAudioFileName, kindFromMime } from "../media/mime.js";
 import { logVerbose, shouldLogVerbose } from "../globals.js";
 import { fetchWithTimeout } from "./providers/shared.js";
 import type { MediaAttachment, MediaUnderstandingCapability } from "./types.js";
@@ -100,23 +100,32 @@ export function normalizeAttachments(ctx: MsgContext): MediaAttachment[] {
   ];
 }
 
-export function isVideoAttachment(attachment: MediaAttachment): boolean {
-  if (attachment.mime?.startsWith("video/")) return true;
+export function resolveAttachmentKind(
+  attachment: MediaAttachment,
+): "image" | "audio" | "video" | "unknown" {
+  const kind = kindFromMime(attachment.mime);
+  if (kind !== "unknown") return kind;
+
   const ext = getFileExtension(attachment.path ?? attachment.url);
-  if (!ext) return false;
-  return [".mp4", ".mov", ".mkv", ".webm", ".avi", ".m4v"].includes(ext);
+  if (!ext) return "unknown";
+  if ([".mp4", ".mov", ".mkv", ".webm", ".avi", ".m4v"].includes(ext)) return "video";
+  if (isAudioFileName(attachment.path ?? attachment.url)) return "audio";
+  if ([".png", ".jpg", ".jpeg", ".webp", ".gif", ".bmp", ".tiff", ".tif"].includes(ext)) {
+    return "image";
+  }
+  return "unknown";
+}
+
+export function isVideoAttachment(attachment: MediaAttachment): boolean {
+  return resolveAttachmentKind(attachment) === "video";
 }
 
 export function isAudioAttachment(attachment: MediaAttachment): boolean {
-  if (attachment.mime?.startsWith("audio/")) return true;
-  return isAudioFileName(attachment.path ?? attachment.url);
+  return resolveAttachmentKind(attachment) === "audio";
 }
 
 export function isImageAttachment(attachment: MediaAttachment): boolean {
-  if (attachment.mime?.startsWith("image/")) return true;
-  const ext = getFileExtension(attachment.path ?? attachment.url);
-  if (!ext) return false;
-  return [".png", ".jpg", ".jpeg", ".webp", ".gif", ".bmp", ".tiff", ".tif"].includes(ext);
+  return resolveAttachmentKind(attachment) === "image";
 }
 
 function isAbortError(err: unknown): boolean {