fix(media): parse lowercase media directives

src/auto-reply/reply/reply-delivery.test.ts

@@ -125,6 +125,20 @@ describe("createBlockReplyDeliveryHandler", () => {
     });
   });
 
+  it("parses lowercase media directives in block replies before path normalization", () => {
+    const normalized = normalizeReplyPayloadDirectives({
+      payload: { text: "media: ./report.pdf" },
+      trimLeadingWhitespace: true,
+      parseMode: "auto",
+    });
+
+    expect(normalized.payload).toMatchObject({
+      text: undefined,
+      mediaUrl: "./report.pdf",
+      mediaUrls: ["./report.pdf"],
+    });
+  });
+
   it("passes normalized media block replies through media path normalization", async () => {
     const blockReplyPipeline = {
       enqueue: vi.fn(),

src/auto-reply/reply/reply-delivery.ts

@@ -26,7 +26,7 @@ export function normalizeReplyPayloadDirectives(params: {
     parseMode === "always" ||
     (parseMode === "auto" &&
       (sourceText.includes("[[") ||
-        sourceText.includes("MEDIA:") ||
+        /media:/i.test(sourceText) ||
         sourceText.includes(silentToken)));
 
   const parsed = shouldParse

src/media/parse.ts

@@ -184,7 +184,7 @@ export function splitMediaFromOutput(raw: string): {
     }
 
     const trimmedStart = line.trimStart();
-    if (!trimmedStart.startsWith("MEDIA:")) {
+    if (!trimmedStart.toUpperCase().startsWith("MEDIA:")) {
       keptLines.push(line);
       pushTextSegment(line);
       lineOffset += line.length + 1; // +1 for newline