fix: correct image result format for OpenAI compatibility to fix #632 (#634)

- Change image result type from 'image' to 'image_url' to match OpenAI API expectations
- Wrap image URL in dict structure: {"url": "..."} instead of plain string
- Update SearchResultPostProcessor to handle dict-based image_url during duplicate removal
- Update tests to validate new image format

This fixes the 400 error: Invalid value: 'image'. Supported values are: 'text', 'image_url'...

Co-authored-by: Willem Jiang <143703838+willem-bd@users.noreply.github.com>

src/tools/search_postprocessor.py

@@ -201,7 +201,14 @@ class SearchResultPostProcessor:
     def _remove_duplicates(self, result: Dict, seen_urls: set) -> Dict:
         """Remove duplicate results"""
 
-        url = result.get("url", result.get("image_url", ""))
+        url = result.get("url")
+        if not url:
+            image_url_val = result.get("image_url", "")
+            if isinstance(image_url_val, dict):
+                url = image_url_val.get("url", "")
+            else:
+                url = image_url_val
+        
         if url and url not in seen_urls:
             seen_urls.add(url)
             return result.copy()  # Return a copy to avoid modifying original

src/tools/tavily_search/tavily_search_api_wrapper.py

@@ -114,8 +114,8 @@ class EnhancedTavilySearchAPIWrapper(OriginalTavilySearchAPIWrapper):
         images = raw_results["images"]
         for image in images:
             clean_result = {
-                "type": "image",
-                "image_url": image["url"],
+                "type": "image_url",
+                "image_url": {"url": image["url"]},
                 "image_description": image["description"],
             }
             clean_results.append(clean_result)

tests/unit/tools/test_tavily_search_api_wrapper.py

@@ -166,8 +166,8 @@ class TestEnhancedTavilySearchAPIWrapper:
 
         # Test image result
         image_result = result[1]
-        assert image_result["type"] == "image"
-        assert image_result["image_url"] == "https://example.com/image.jpg"
+        assert image_result["type"] == "image_url"
+        assert image_result["image_url"] == {"url": "https://example.com/image.jpg"}
         assert image_result["image_description"] == "Test image description"
 
     def test_clean_results_without_raw_content(self, wrapper):