File size: 5,240 Bytes
d6d88ae
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
import os

from dotenv import load_dotenv
load_dotenv() # Loads variables from .env into the environment

# os.environ["LANGFUSE_HOST"] = os.getenv("LANGFUSE_HOST")
# os.environ["LANGFUSE_PUBLIC_KEY"] = os.getenv("LANGFUSE_PUBLIC_KEY")
# os.environ["LANGFUSE_SECRET_KEY"] = os.getenv("LANGFUSE_SECRET_KEY")

# os.environ["HF_TOKEN"] = os.getenv("HF_token")
# os.environ["GEMINI_API_KEY"] = os.getenv("GEMINI_API_KEY")
#os.environ["TAVILY_API_KEY"] = os.getenv("TRAVILY_SECRET_KEY")

from langfuse import get_client
langfuse = get_client()
# Verify connection
if langfuse.auth_check():
    print("Langfuse client is authenticated and ready!")
else:
    print("Authentication failed. Please check your credentials and host.")


from smolagents import OpenAIServerModel
gemini_model = OpenAIServerModel(
    model_id="gemini-2.5-flash",
    api_base="https://generativelanguage.googleapis.com/v1beta/openai/",
    api_key=os.getenv("GEMINI_API_KEY")
)

# from smolagents import TransformersModel
# VL_model = TransformersModel(
#     model_id="Qwen/Qwen2.5-VL-32B-Instruct",
#     device_map="auto",
#     torch_dtype="auto",
# )

# coder_model = TransformersModel(
#     model_id="Qwen/Qwen2.5-Coder-32B-Instruct",
#     device_map="auto",
#     torch_dtype="auto",
# )


#from Tools.searchtool import WebSearchTool
from Tools.mywebpagevisit import CustomVisitWebpageTool
from smolagents import (
    CodeAgent,
    WebSearchTool,
    #WikipediaSearchTool
)

web_search = WebSearchTool()
visit_webpage = CustomVisitWebpageTool()
#wikipedia_search = WikipediaSearchTool()

from openinference.instrumentation.smolagents import SmolagentsInstrumentor
SmolagentsInstrumentor().instrument()

web_agent = CodeAgent(
    tools=[WebSearchTool,CustomVisitWebpageTool],
    model= gemini_model,
    additional_authorized_imports = ['requests', 'bs4','json'],
    name="web_agent",
    description="This is an agent that can do web search using 'WebSearchTool' and visit a webpage using 'CustomVisitWebpageTool'."
)
 
from Tools.ytdownload import youtube_audio_downloader,youtube_video_downloader

yt_agent = CodeAgent(
    tools=[youtube_audio_downloader,youtube_video_downloader],
    model= gemini_model,
    name="youtube_downloader_agent",
    description="This is an agent that can download youtube audio file using 'youtube_audio_downloader'and youtube video file using 'youtube_video_downloader'."
)

from Tools.transcriber import transcriber

transcriber_agent = CodeAgent(
    tools=[transcriber],
    model= gemini_model,
    name="transcriber_agent",
    description="This is an agent that can transcribe a local audio file using 'transcriber'.\n\nDo not remove adjectives or descriptors of ingredients if not specifically mentioned."
)

from Tools.visual_reasoner import video_reasoner, image_reasoner
VL_agent = CodeAgent(
    tools=[video_reasoner,image_reasoner],
    model= gemini_model,
    name="VL_agent",
    description="This is an agent that can answer questions on a video using 'video_reasoner' and on a image using 'image_reasoner'."
)

excel_agent = CodeAgent(
    tools=[],
    model= gemini_model,
    name="excel_agent",
    description="This is an agent that can work with excel.",
    add_base_tools=False,
    additional_authorized_imports = ['pandas', 'numpy', 'math']
)

from Tools.code_reader import code_reader
manager_agent = CodeAgent(
    tools=[code_reader],
    model=gemini_model,
    managed_agents=[web_agent, yt_agent, transcriber_agent, VL_agent],
    additional_authorized_imports=[
        "numpy","os","pandas",'scipy','regex','math'
    ],
    planning_interval=5,
    verbosity_level=2,
    max_steps=10,
    description= "You are a general AI assistant. I will ask you a question. Report your thoughts, and finish your answer with the following template: FINAL ANSWER: [YOUR FINAL ANSWER]. YOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings. If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise. If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise. If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string."
)

# manager_agent.run(
#     "How many studio albums were published by Mercedes Sosa between 2000 and 2009 (included)? You can use the latest 2022 version of english wikipedia."
# )

class MyAgent:
    def __init__(self):
        print("BasicAgent initialized.")

    def __call__(self, question: str) -> str:
        print(f"Agent received question (first 50 chars): {question[:50]}...")
        try:
            answer = manager_agent.run(question)
            return answer
        except Exception as e:
            error = f"An error occurred while processing the question: {e}"
            print(error)
            return error


# if __name__ == "__main__":
#     question = """
# What was the actual enrollment of the Malko competition in 2023?
# """
#     agent = MyAgent()
#     answer = agent(question)
#     print(f"Answer: {answer}")