Webpage MCP Server

================================================================================ Running 3 test cases... ================================================================================ Starting test: github_search ================================================================================ INFO [github_search] ================================================================================ TEST: github_search INFO [github_search] TEST: github_search ================================================================================ INFO [github_search] ================================================================================ INFO [github_search] Action: search for repositories on GitHub INFO [github_search] Action: search for repositories on GitHub Website: https://github.com INFO [github_search] Website: https://github.com INFO [github_search] Running action_agent... INFO [github_search] Running action_agent... ✓ Function: search_github_repositories INFO [github_search] ✓ Function: search_github_repositories INFO [github_search] Running input_agent... INFO [github_search] Running input_agent... ✓ Generated 5 test inputs INFO [github_search] ✓ Generated 5 test inputs INFO [github_search] Running task_agent... INFO [github_search] Running task_agent... ✓ Task description created INFO [github_search] ✓ Task description created INFO [github_search] Running browser_loop... INFO [github_search] Running browser_loop... INFO [github_search] Testing: {'query': 'machine learning'} Attempt 1/3 INFO [service] Using anonymized telemetry, see https://docs.browser-use.com/development/telemetry. INFO [Agent] 🚀 Task: TASK DESCRIPTION: Objective: The goal is to search for repositories on GitHub using a specific query term, and extract relevant information about each repository that matches the search. This will involve interacting with the GitHub website through its search functionality. Data to Extract: For each repository that appears in the search results, the following details need to be extracted: 1. **Name**: The name of the repository. 2. **URL**: The direct link to the repository on GitHub. 3. **Description**: A brief description of the repository, which provides insights into its purpose and functionality. Process Overview: 1. Start by navigating to the GitHub homepage. 2. Use the search functionality to enter the provided query term. 3. Wait for the search results page to load. 4. Locate the section that displays the list of repositories matching the query. 5. For each repository listed in the search results, collect the following information as specified: - **Name** of the repository, which is typically a clickable link. - **URL** which is the hyperlink associated with the repository name. - **Description** which is usually found just below the name of the repository. Final Output: The final output should be a structured list of repositories, where each repository is represented as a dictionary containing the following keys and their corresponding values: - 'name': The name of the repository (string) - 'url': The URL of the repository (string) - 'description': A brief description of the repository (string) The output format should be a list of these dictionaries, allowing for easy review and reference of all the found repositories based on the search query. THESE ARE YOUR INPUTS: - query: machine learning  INFO [service] ──────────────────────────────────────── INFO [service] 🔐 To view this run in Browser Use Cloud, authenticate with: INFO [service] 👉 browser-use auth INFO [service] or: python -m browser_use.cli auth INFO [service] ──────────────────────────────────────── INFO [Agent] 🧠 Starting a browser-use agent with version 0.8.0 and model=gpt-4o-mini INFO [Agent] INFO [Agent] 📍 Step 1: INFO [Agent] ❔ Eval: The page is empty and no interactive elements are available yet. Verdict: Uncertain. INFO [Agent] 🎯 Next goal: Wait for a few seconds to allow the GitHub homepage to load fully and then check for interactive elements. INFO [Agent] 🦾 [ACTION 1/1] wait: seconds: 5 INFO [tools] 🕒 waited for 2 seconds + 3 seconds for LLM call INFO [Agent] INFO [Agent] 📍 Step 2: INFO [Agent] ❔ Eval: The page is still empty, indicating that it has not fully loaded yet. Verdict: Uncertain. INFO [Agent] 🎯 Next goal: Wait for a few more seconds to allow the GitHub homepage to load fully and then check for interactive elements. INFO [Agent] 🦾 [ACTION 1/1] wait: seconds: 5 INFO [tools] 🕒 waited for 2 seconds + 3 seconds for LLM call INFO [Agent] INFO [Agent] 📍 Step 3: INFO [Agent] 👍 Eval: The page is still empty after waiting multiple times, indicating that it has not loaded successfully. Verdict: Uncertain. INFO [Agent] 🎯 Next goal: Wait for an additional few seconds to see if the GitHub homepage loads completely or consider refreshing the page if it remains unresponsive. INFO [Agent] 🦾 [ACTION 1/1] wait: seconds: 5 INFO [tools] 🕒 waited for 2 seconds + 3 seconds for LLM call INFO [Agent] INFO [Agent] 📍 Step 4: INFO [Agent] 👍 Eval: The page remains empty after multiple waits, indicating that it has not loaded successfully. Verdict: Uncertain. INFO [Agent] 🎯 Next goal: Refresh the GitHub homepage to attempt loading it again. INFO [Agent] 🦾 [ACTION 1/1] navigate: url: https://github.com/, new_tab: False INFO [tools] 🔗 Navigated to https://github.com/ INFO [Agent] INFO [Agent] 📍 Step 5: INFO [Agent] ❔ Eval: The page has loaded, but I was unable to find the search bar or any interactive elements for searching. Verdict: Uncertain. INFO [Agent] 🎯 Next goal: Scroll down the page to see if more interactive elements, including the search bar, become available. INFO [Agent] 🦾 [ACTION 1/1] scroll: down: True, pages: 1.0, index: None INFO [tools] 🔍 Scrolled down the page by one page (469px) INFO [Agent] INFO [Agent] 📍 Step 6: INFO [Agent] ❔ Eval: I attempted to scroll down the page to find interactive elements, but the search bar is still not visible. Verdict: Uncertain. INFO [Agent] 🎯 Next goal: Continue scrolling down the page to see if the search bar becomes available for entering the query 'machine learning'. INFO [Agent] 🦾 [ACTION 1/1] scroll: down: True, pages: 1.0, index: None INFO [tools] 🔍 Scrolled down the page by one page (469px) INFO [Agent] INFO [Agent] 📍 Step 7: INFO [Agent] ❔ Eval: I attempted to scroll down the page to find interactive elements, but the search bar is still not visible. Verdict: Uncertain. INFO [Agent] 🎯 Next goal: Scroll down the page again to see if the search bar becomes available for entering the query 'machine learning'. INFO [Agent] 🦾 [ACTION 1/1] scroll: down: True, pages: 1.0, index: None INFO [tools] 🔍 Scrolled down the page by one page (469px) INFO [Agent] INFO [Agent] 📍 Step 8: INFO [Agent] ❔ Eval: I attempted to scroll down the page to find interactive elements, but the search bar is still not visible. Verdict: Uncertain. INFO [Agent] 🎯 Next goal: Scroll down the page again to see if the search bar becomes available for entering the query 'machine learning'. INFO [Agent] 🦾 [ACTION 1/1] scroll: down: True, pages: 1.0, index: None INFO [tools] 🔍 Scrolled down the page by one page (469px) INFO [Agent] INFO [Agent] 📍 Step 9: INFO [Agent] ❔ Eval: I attempted to scroll down the page multiple times but was unable to find the search bar. Verdict: Uncertain. INFO [Agent] 🎯 Next goal: Scroll down the page again to see if the search bar becomes available for entering the query 'machine learning'. INFO [Agent] 🦾 [ACTION 1/1] scroll: down: True, pages: 1.0, index: None INFO [tools] 🔍 Scrolled down the page by one page (469px) INFO [Agent] INFO [Agent] 📍 Step 10: INFO [Agent] 👍 Eval: The attempt to locate the search bar on GitHub was unsuccessful, resulting in no progress towards completing the user request. INFO [Agent] 🎯 Next goal: Since I could not complete the task of searching for repositories, I will finalize my findings and indicate that the task is incomplete. INFO [Agent] 🦾 [ACTION 1/1] text="The task of searching for repositories related to 'machine learning' on GitHub could not be completed as I was unable to locate the search bar or any interactive elements necessary for performing the search. Therefore, this task remains unfinished." success=False files_to_display=None INFO [Agent] 📄  Final Result: The task of searching for repositories related to 'machine learning' on GitHub could not be completed as I was unable to locate the search bar or any interactive elements necessary for performing the search. Therefore, this task remains unfinished. INFO [Agent] ❌ Task completed without success WARNING [cdp_use.client] WebSocket connection closed: no close frame received or sent Traceback (most recent call last): File "/Users/kushbhuwalka/Documents/projects/dedalus-mcp-2/test_cases.py", line 128, in <module> asyncio.run(main()) ~~~~~~~~~~~^^^^^^^^ File "/opt/homebrew/Cellar/python@3.13/3.13.1/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/runners.py", line 194, in run return runner.run(main) ~~~~~~~~~~^^^^^^ File "/opt/homebrew/Cellar/python@3.13/3.13.1/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/runners.py", line 118, in run return self._loop.run_until_complete(task) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^ File "/opt/homebrew/Cellar/python@3.13/3.13.1/Frameworks/Python.framework/Versions/3.13/lib/python3.13/asyncio/base_events.py", line 720, in run_until_complete return future.result() ~~~~~~~~~~~~~^^ File "/Users/kushbhuwalka/Documents/projects/dedalus-mcp-2/test_cases.py", line 110, in main await run_test( ...<3 lines>... ) File "/Users/kushbhuwalka/Documents/projects/dedalus-mcp-2/test_cases.py", line 70, in run_test state = await browser_loop(state) ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/kushbhuwalka/Documents/projects/dedalus-mcp-2/agents/browser_loop.py", line 20, in browser_loop evaluation = await scoring_agent(current_task, str(browser_result)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/kushbhuwalka/Documents/projects/dedalus-mcp-2/agents/browser_agent.py", line 61, in scoring_agent json_data = json.loads(eval_str) File "/opt/homebrew/Cellar/python@3.13/3.13.1/Frameworks/Python.framework/Versions/3.13/lib/python3.13/json/__init__.py", line 346, in loads return _default_decoder.decode(s) ~~~~~~~~~~~~~~~~~~~~~~~^^^ File "/opt/homebrew/Cellar/python@3.13/3.13.1/Frameworks/Python.framework/Versions/3.13/lib/python3.13/json/decoder.py", line 345, in decode obj, end = self.raw_decode(s, idx=_w(s, 0).end()) ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/homebrew/Cellar/python@3.13/3.13.1/Frameworks/Python.framework/Versions/3.13/lib/python3.13/json/decoder.py", line 363, in raw_decode raise JSONDecodeError("Expecting value", s, err.value) from None json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)