Implement Examples from AppAgent and Verify If It Works

[EugeneVuong]

1. Go to Find a Task in the AppAgent
2. Copy the State Prompt of the Said Task
3. Create a New File with the same name as the task in the NetGent Repo (Ensure that the file is in the right folder (conference, streaming, and browsing)

import json
from netgent import NetGent, StatePrompt
from langchain_google_vertexai import ChatVertexAI
from langchain_google_genai import ChatGoogleGenerativeAI
from dotenv import load_dotenv
load_dotenv()
agent = NetGent(llm=ChatVertexAI(model="gemini-2.0-flash-exp", temperature=0.2), llm_enabled=True, user_data_dir="examples/user_data")
prompt = [
        <Add State Prompt Here>
    ]

try:
    with open("examples/<Task Label>/results/<Name of Task>_result.json", "r") as f:
        result = json.load(f)
except FileNotFoundError:
    result = []

result = agent.run(state_prompts=prompt, state_repository=result)

input("Press Enter to continue...")
with open("examples/<Task Label>/results/<Name of Task>.json", "w") as f:
    json.dump(result["state_repository"], f, indent=2)

5. Run the action to see if it works and generate the result JSON in the correct folder.

[jaber-the-great]

@sumedh-aerram I reviewed the workflows and identified the following required changes:

1. Move all workflow files to the examples directory.
2. Remove all UCSB-related credentials, including the SNL client username and streaming platform passwords.
3. Add a brief comment at the top of each file (two to three sentences) describing the purpose and function of the workflow.

[jaber-the-great]

The json files are not created in the example workflows. Make sure:

1. The workflow correctly generates the code
2. You can execute the generated code without using agents
3. There are no credentials in the prompts and json files you commit to the repo. Edit the README file and explain the users need to replace that part (credentials) in order to run the existing workflows or use the prompts that we have made.