update

app.py

@@ -370,6 +370,9 @@ class BasicAgent:
                 "openpyxl",
                 "json",
                 "wikipediaapi",
+                "pytube",
+                "pytubefix",
+                "pytubefix.cli",
             ],
             name="multimodal_agent",
             description="""
@@ -409,6 +412,9 @@ class BasicAgent:
                 "openpyxl",
                 "json",
                 "wikipediaapi",
+                "pytube",
+                "pytubefix",
+                "pytubefix.cli",
             ],
             planning_interval=2,
             max_steps=10,
@@ -425,15 +431,8 @@ class BasicAgent:
                 - Please answer as concisely as possible.
                 - If the question asks for a number, please return a numerical answer without unit (unless unit is specifically asked for). For example: 3 instead of three, 0 instead of None, 3 instead of $3.
                 - If the question asks for a number with specific decimal places, please format the number into string with the same decimal places. For example: 3.00 instead of 3.
-                - To get table data from Wikipedia or any HTML page content, please follow the following steps:
-                    - 1. Identify the section you want to extract.
-                    - 2.Use `pandas.read_html` to extract the table data from that section.
                 - If the question asks for a list, please make sure that the elements are separated by a comma(`,`) and a space(` `). For example: `1, 2, 3` instead of `1,2,3`.
                 - If the question asks for name without abbreviations, please ALWAYS ask `multimodal_agent` for the FULL name of final answer to ensure NO abbreviation is included in Final Answer. For example: `United States` instead of `US`.
-                - Just give up and return a guess answer if the request requires you to do any of the following:
-                    - Watch Youtube videos.
-                    - Convert Youtube videos to audio.
-                    - Solve chess puzzle from an image.
         """
         if "food" in question.lower() or "drink" in question.lower():
             prompt = f"""

requirements.txt

@@ -16,3 +16,5 @@ openai-whisper
 openpyxl
 youtube-transcript-api
 wikipedia-api
+pytube
+pytubefix