feat(api): support file uploads via JSON base64 and multipart/form-data

tests/test_openai_api.py

@@ -194,6 +194,7 @@ async def test_successful_request_uses_fixed_api_session(aiohttp_client, mock_ag
     assert body["model"] == "test-model"
     mock_agent.process_direct.assert_called_once_with(
         content="hello",
+        media=None,
         session_key=API_SESSION_KEY,
         channel="api",
         chat_id=API_CHAT_ID,
@@ -205,7 +206,7 @@ async def test_successful_request_uses_fixed_api_session(aiohttp_client, mock_ag
 async def test_followup_requests_share_same_session_key(aiohttp_client) -> None:
     call_log: list[str] = []
 
-    async def fake_process(content, session_key="", channel="", chat_id=""):
+    async def fake_process(content, session_key="", channel="", chat_id="", **kwargs):
         call_log.append(session_key)
         return f"reply to {content}"
 
@@ -236,7 +237,7 @@ async def test_followup_requests_share_same_session_key(aiohttp_client) -> None:
 async def test_fixed_session_requests_are_serialized(aiohttp_client) -> None:
     order: list[str] = []
 
-    async def slow_process(content, session_key="", channel="", chat_id=""):
+    async def slow_process(content, session_key="", channel="", chat_id="", **kwargs):
         order.append(f"start:{content}")
         await asyncio.sleep(0.1)
         order.append(f"end:{content}")
@@ -307,12 +308,12 @@ async def test_multimodal_content_extracts_text(aiohttp_client, mock_agent) -> N
         },
     )
     assert resp.status == 200
-    mock_agent.process_direct.assert_called_once_with(
-        content="describe this",
-        session_key=API_SESSION_KEY,
-        channel="api",
-        chat_id=API_CHAT_ID,
-    )
+    call_kwargs = mock_agent.process_direct.call_args.kwargs
+    assert call_kwargs["content"] == "describe this"
+    assert call_kwargs["session_key"] == API_SESSION_KEY
+    assert call_kwargs["channel"] == "api"
+    assert call_kwargs["chat_id"] == API_CHAT_ID
+    assert len(call_kwargs.get("media") or []) >= 0  # base64 images saved to disk
 
 
 @pytest.mark.skipif(not HAS_AIOHTTP, reason="aiohttp not installed")
@@ -320,7 +321,7 @@ async def test_multimodal_content_extracts_text(aiohttp_client, mock_agent) -> N
 async def test_empty_response_retry_then_success(aiohttp_client) -> None:
     call_count = 0
 
-    async def sometimes_empty(content, session_key="", channel="", chat_id=""):
+    async def sometimes_empty(content, session_key="", channel="", chat_id="", **kwargs):
         nonlocal call_count
         call_count += 1
         if call_count == 1:
@@ -351,7 +352,7 @@ async def test_empty_response_falls_back(aiohttp_client) -> None:
 
     call_count = 0
 
-    async def always_empty(content, session_key="", channel="", chat_id=""):
+    async def always_empty(content, session_key="", channel="", chat_id="", **kwargs):
         nonlocal call_count
         call_count += 1
         return ""
@@ -371,3 +372,31 @@ async def test_empty_response_falls_back(aiohttp_client) -> None:
     body = await resp.json()
     assert body["choices"][0]["message"]["content"] == EMPTY_FINAL_RESPONSE_MESSAGE
     assert call_count == 2
+
+
+@pytest.mark.asyncio
+async def test_process_direct_accepts_media() -> None:
+    """process_direct should forward media paths to _process_message."""
+    from nanobot.agent.loop import AgentLoop
+
+    loop = AgentLoop.__new__(AgentLoop)
+    loop._connect_mcp = AsyncMock()
+
+    captured_msg = None
+
+    async def fake_process(msg, *, session_key="", on_progress=None, on_stream=None, on_stream_end=None):
+        nonlocal captured_msg
+        captured_msg = msg
+        return None
+
+    loop._process_message = fake_process
+
+    await loop.process_direct(
+        content="analyze this",
+        media=["/tmp/image.png", "/tmp/report.pdf"],
+        session_key="test:1",
+    )
+
+    assert captured_msg is not None
+    assert captured_msg.media == ["/tmp/image.png", "/tmp/report.pdf"]
+    assert captured_msg.content == "analyze this"