fix(vision): use MiniMax type="video" block (not input_video) + tests

The salvaged conversion emitted type:"input_video", which MiniMax M3 rejects
just like the original video_url block. Per MiniMax's Anthropic-compat docs,
the video content block is type:"video" with an image-style source (base64 or
url). Fixes the block type, converts URL-based videos too, and adds 4 video
conversion tests (none shipped with the original PR).

agent/auxiliary_client.py

@@ -4756,11 +4756,11 @@ def _is_anthropic_compat_endpoint(provider: str, base_url: str) -> bool:
 
 
 def _convert_openai_images_to_anthropic(messages: list) -> list:
-    """Convert OpenAI ``image_url`` and ``video_url`` content blocks to Anthropic format.
+    """Convert OpenAI ``image_url``/``video_url`` blocks to Anthropic format.
 
     Converts:
-    - ``image_url`` blocks → Anthropic ``image`` blocks
-    - ``video_url`` blocks → Anthropic ``input_video`` blocks
+    - ``image_url`` blocks to Anthropic ``image`` blocks
+    - ``video_url`` blocks to Anthropic ``video`` blocks (MiniMax M3 compat)
 
     Only touches messages that have list-type content with ``image_url`` or
     ``video_url`` blocks; plain text messages pass through unchanged.
@@ -4801,6 +4801,13 @@ def _convert_openai_images_to_anthropic(messages: list) -> list:
                     })
                 changed = True
             elif block.get("type") == "video_url":
+                # MiniMax's Anthropic-compatible endpoint expects a "video"
+                # block (not OpenAI's "video_url", and not "input_video").
+                # See https://platform.minimax.io/docs/api-reference/text-anthropic-api
+                # — the Messages-field table lists type="video" (M3 only,
+                # URL/base64/mm_file://). The source shape mirrors the "image"
+                # block: base64 → {type:"base64", media_type, data}, URL →
+                # {type:"url", url}.
                 video_url_val = (block.get("video_url") or {}).get("url", "")
                 if video_url_val.startswith("data:"):
                     # Parse data URI: data:<media_type>;base64,<data>
@@ -4809,7 +4816,7 @@ def _convert_openai_images_to_anthropic(messages: list) -> list:
                     if ":" in header and ";" in header:
                         media_type = header.split(":", 1)[1].split(";", 1)[0]
                     new_content.append({
-                        "type": "input_video",
+                        "type": "video",
                         "source": {
                             "type": "base64",
                             "media_type": media_type,
@@ -4817,10 +4824,14 @@ def _convert_openai_images_to_anthropic(messages: list) -> list:
                         },
                     })
                 else:
-                    # URL-based video — Anthropic protocol doesn't have a
-                    # native URL-based video block; pass through as-is and
-                    # let the provider handle (or reject) it.
-                    new_content.append(block)
+                    # URL-based video
+                    new_content.append({
+                        "type": "video",
+                        "source": {
+                            "type": "url",
+                            "url": video_url_val,
+                        },
+                    })
                 changed = True
             else:
                 new_content.append(block)

tests/agent/test_auxiliary_client.py

@@ -2178,6 +2178,61 @@ class TestAnthropicCompatImageConversion:
         result = _convert_openai_images_to_anthropic(messages)
         assert result[0]["content"][0]["source"]["media_type"] == "image/jpeg"
 
+    def test_base64_video_converted_to_video_block(self):
+        # MiniMax M3's Anthropic-compatible endpoint expects type="video"
+        # (not OpenAI's "video_url", not "input_video").
+        from agent.auxiliary_client import _convert_openai_images_to_anthropic
+        messages = [{
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "What happens in this clip?"},
+                {"type": "video_url", "video_url": {"url": "data:video/mp4;base64,AAAA"}},
+            ],
+        }]
+        result = _convert_openai_images_to_anthropic(messages)
+        vid_block = result[0]["content"][1]
+        assert vid_block["type"] == "video"
+        assert vid_block["source"]["type"] == "base64"
+        assert vid_block["source"]["media_type"] == "video/mp4"
+        assert vid_block["source"]["data"] == "AAAA"
+
+    def test_video_media_type_parsed_from_data_uri(self):
+        from agent.auxiliary_client import _convert_openai_images_to_anthropic
+        messages = [{
+            "role": "user",
+            "content": [
+                {"type": "video_url", "video_url": {"url": "data:video/quicktime;base64,QQ=="}}
+            ],
+        }]
+        result = _convert_openai_images_to_anthropic(messages)
+        assert result[0]["content"][0]["source"]["media_type"] == "video/quicktime"
+
+    def test_url_video_converted_to_video_block(self):
+        from agent.auxiliary_client import _convert_openai_images_to_anthropic
+        messages = [{
+            "role": "user",
+            "content": [
+                {"type": "video_url", "video_url": {"url": "https://example.com/clip.mp4"}}
+            ],
+        }]
+        result = _convert_openai_images_to_anthropic(messages)
+        vid_block = result[0]["content"][0]
+        assert vid_block["type"] == "video"
+        assert vid_block["source"] == {"type": "url", "url": "https://example.com/clip.mp4"}
+
+    def test_mixed_image_and_video_both_converted(self):
+        from agent.auxiliary_client import _convert_openai_images_to_anthropic
+        messages = [{
+            "role": "user",
+            "content": [
+                {"type": "image_url", "image_url": {"url": "data:image/png;base64,iVBOR"}},
+                {"type": "video_url", "video_url": {"url": "data:video/mp4;base64,AAAA"}},
+            ],
+        }]
+        result = _convert_openai_images_to_anthropic(messages)
+        assert result[0]["content"][0]["type"] == "image"
+        assert result[0]["content"][1]["type"] == "video"
+
 
 class _AuxAuth401(Exception):
     status_code = 401