From d30c4d00d3bef0e85fd2e92d1deb0ac4c696d6c5 Mon Sep 17 00:00:00 2001
From: Willem Jiang <willem.jiang@gmail.com>
Date: Tue, 21 Oct 2025 10:00:33 +0800
Subject: [PATCH] fix: convert crawl_tool dict return to JSON string for type
 consistency (#636)

Keep fixing #631
This pull request updates the crawl_tool function to return its results as a JSON string instead of a dictionary, and adjusts the unit tests accordingly to handle the new return type. The changes ensure consistent serialization of output and proper validation in tests.
---
 src/tools/crawl.py             |  3 ++-
 tests/unit/tools/test_crawl.py | 13 ++++++++-----
 2 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/src/tools/crawl.py b/src/tools/crawl.py
index 552e813..fbb349e 100644
--- a/src/tools/crawl.py
+++ b/src/tools/crawl.py
@@ -1,6 +1,7 @@
 # Copyright (c) 2025 Bytedance Ltd. and/or its affiliates
 # SPDX-License-Identifier: MIT
 
+import json
 import logging
 from typing import Annotated
 
@@ -22,7 +23,7 @@ def crawl_tool(
     try:
         crawler = Crawler()
         article = crawler.crawl(url)
-        return {"url": url, "crawled_content": article.to_markdown()[:1000]}
+        return json.dumps({"url": url, "crawled_content": article.to_markdown()[:1000]})
     except BaseException as e:
         error_msg = f"Failed to crawl. Error: {repr(e)}"
         logger.error(error_msg)
diff --git a/tests/unit/tools/test_crawl.py b/tests/unit/tools/test_crawl.py
index f43b11c..a21c426 100644
--- a/tests/unit/tools/test_crawl.py
+++ b/tests/unit/tools/test_crawl.py
@@ -1,3 +1,4 @@
+import json
 from unittest.mock import Mock, patch
 
 from src.tools.crawl import crawl_tool
@@ -21,10 +22,11 @@ class TestCrawlTool:
         result = crawl_tool(url)
 
         # Assert
-        assert isinstance(result, dict)
-        assert result["url"] == url
-        assert "crawled_content" in result
-        assert len(result["crawled_content"]) <= 1000
+        assert isinstance(result, str)
+        result_dict = json.loads(result)
+        assert result_dict["url"] == url
+        assert "crawled_content" in result_dict
+        assert len(result_dict["crawled_content"]) <= 1000
         mock_crawler_class.assert_called_once()
         mock_crawler.crawl.assert_called_once_with(url)
         mock_article.to_markdown.assert_called_once()
@@ -45,7 +47,8 @@ class TestCrawlTool:
         result = crawl_tool(url)
 
         # Assert
-        assert result["crawled_content"] == short_content
+        result_dict = json.loads(result)
+        assert result_dict["crawled_content"] == short_content
 
     @patch("src.tools.crawl.Crawler")
     @patch("src.tools.crawl.logger")