communication between two uAgents and Flask API #494

Data-Scientist-Sahil · 2024-08-09T07:33:34Z

Data-Scientist-Sahil
Aug 9, 2024

We are stuck on this problem and therefore unable to integrate our backend and frontend services and the submission of hackathon round 1 is stuck. Fetach AI X IIT Guwahati.
When we hit the code with app.py it is not able to accept the message from flask server but the secondary uagent quote.py is working and even sending the message. We are confused.

this is my app.py

from flask import Flask, render_template, request, jsonify
import asyncio
from uagents import Agent, Context, Model
from uagents.query import query
from uagents.envelope import Envelope
import json

app = Flask(__name__)

# Model classes for the agent
class Quote(Model):
    feeling: str
    context: str

class Response(Model):
    response: str

# Hardcoded address of the quote agent
quote_address = "agent1q249x2snexu4f6am4rnk4e8dmuw9sqj7x886e7ydp35rnuqqgf9pu5p7yev"

# Route for the main page
@app.route('/')
def index():
    return render_template('index.html')

# Route for handling user input and getting response from the agent
@app.route('/get_quote', methods=['POST'])
def get_quote():
    feeling = request.form['feeling']
    context = request.form.get('context', "")
    
    # Sending and receiving response from the agent
    loop = asyncio.new_event_loop()
    asyncio.set_event_loop(loop)
    response = loop.run_until_complete(query_agent(feeling, context))
    
    # Debugging statement to print the response object
    print(f"Response received: {response}")

    # Check if the response has the expected structure
    # if isinstance(response, dict) and 'response' in response:
    #     context += f"Feeling: {feeling}\nCoach: {response['response']}\n"
    #     return jsonify({'response': response['response'], 'context': context})
    # else:
    return jsonify({'response': response['response'], 'context': ""})

# Function to interact with the agent
async def query_agent(feeling, context):
    response = await query(destination=quote_address, message=Quote(feeling=feeling, context=context), timeout=300.0)
    print(response)
    
    # data = json.loads(response.decode_payload())
    # print(data)
    # return data
    
    # data = json.loads(response.decode_payload())
    # return data

if __name__ == '__main__':
    app.run(debug=True, port=8004)

this is my quote.py

import os

from uagents import Agent, Context, Model


class Quote(Model):
    feeling: str  # Represents the user's feeling or problem
    context: str  # Represents the context of the conversation, if any

class Response(Model):
    response: str  # Represents the response from the coach
quote = Agent(
    name="quote",
    port=8001,
    seed="QuoteAgent - Team Clutch",
    endpoint=["http://127.0.0.1:8001/submit"]
)

# Print the agent's address for reference
print(quote.address)

@quote.on_query(model=Quote)
async def handle_message(ctx: Context, sender: str, msg: Quote):
    # Define the prompt template without context
    ctx.logger.info(f"Received message from {sender}: {msg}")
    
    coach_response = "I am working"
    ctx.logger.info(f"Coach response: {coach_response}")
    
    try:
        # Send the coach's response back to the sender
        await ctx.send(sender, Response(response=coach_response))
    except Exception as e:
        ctx.logger.error(f"Failed to send response: {e}")


# Run the agent
if __name__ == "__main__":
    quote.run()

so after i run the quote.py and app.py , the logging information from quote.py informs that the message was received and the message was sent back to app.py , but the logging information from app.py says that initial user request never reached the app.py

Loggings of quote.py
-> INFO: [quote]: Almanac registration is up to date!
INFO: [quote]: Starting server on http://0.0.0.0:8001 (Press CTRL+C to quit)
INFO: [quote]: recieved message : user1nu5nm9uvd9z3zd92k9ql00dvnf4kkpdkn3f99423d527j57tseusefvjam
INFO: [quote]: I am working

Loggings of app.py
-> ERROR: [dispenser]: Failed to deliver message to agent1q249x2snexu4f6am4rnk4e8dmuw9sqj7x886e7ydp35rnuqqgf9pu5p7yev @ ['http://127.0.0.1:8001/submit']: ['Failed to send message: Envelope signature is missing']
MsgStatus(status=<DeliveryStatus.FAILED: 'failed'>, detail='Message delivery failed', destination='agent1q249x2snexu4f6am4rnk4e8dmuw9sqj7x886e7ydp35rnuqqgf9pu5p7yev', endpoint='', session=UUID('3e8df0a8-85c2-4cef-9044-ac4e37b1b4aa'))
Response received: None
INFO:werkzeug:127.0.0.1 - - [09/Aug/2024 12:52:43] "POST /get_quote HTTP/1.1" 500 -
(traceback call info)
return jsonify({'response': response['response'], 'context': ""})
TypeError: 'NoneType' object is not subscriptable

please help us asap via connecting with us.

kshipra-fetch · 2024-08-09T10:14:37Z

kshipra-fetch
Aug 9, 2024
Collaborator

Hey @Data-Scientist-Sahil,

The code below is working fine for us, please make sure the version of the uagents library is 0.14.0.

pip install uagents==0.14.0

from flask import Flask, render_template, request, jsonify
import asyncio
from uagents import Agent, Context, Model
from uagents.query import query
from uagents.envelope import Envelope
import json

app = Flask(__name__)

# Model classes for the agent
class Quote(Model):
    feeling: str
    context: str

class Response(Model):
    response: str

# Hardcoded address of the quote agent
quote_address = "agent1q249x2snexu4f6am4rnk4e8dmuw9sqj7x886e7ydp35rnuqqgf9pu5p7yev"

# Route for the main page
@app.route('/')
def index():
    return render_template('index.html')

# Route for handling user input and getting response from the agent
@app.route('/get_quote', methods=['POST'])
def get_quote():
    feeling = request.form['feeling']
    context = request.form.get('context', "")
    
    # Sending and receiving response from the agent
    loop = asyncio.new_event_loop()
    asyncio.set_event_loop(loop)
    response = loop.run_until_complete(query_agent(feeling, context))
    
    # Debugging statement to print the response object
    print(f"Response received: {response}")

    # Check if the response has the expected structure
    if isinstance(response, dict) and 'response' in response:
        return jsonify({'response': response['response'], 'context': context})
    else:
        return jsonify({'response': 'No valid response received', 'context': ""})

# Function to interact with the agent
async def query_agent(feeling, context):
    response = await query(destination=quote_address, message=Quote(feeling=feeling, context=context), timeout=15.0)
    
    if isinstance(response, Envelope):
        data = json.loads(response.decode_payload())
        return data
    return {"response": "Error occurred while querying the agent"}

if __name__ == '__main__':
    app.run(debug=True, port=8004)

Here are the logs:

INFO:werkzeug: * Restarting with stat
WARNING:werkzeug: * Debugger is active!
INFO:werkzeug: * Debugger PIN: 590-607-864
INFO:werkzeug: * Detected change in '/Users/abhimanyugangani/Desktop/Fetch.AI/integrations/hackathon check/main.py', reloading
INFO:werkzeug: * Restarting with stat
WARNING:werkzeug: * Debugger is active!
INFO:werkzeug: * Debugger PIN: 590-607-864
INFO:werkzeug:127.0.0.1 - - [09/Aug/2024 10:52:10] "GET / HTTP/1.1" 200 -
Response received: {'response': 'I am working'}
INFO:werkzeug:127.0.0.1 - - [09/Aug/2024 10:52:14] "POST /get_quote HTTP/1.1" 200 -
INFO:werkzeug:127.0.0.1 - - [09/Aug/2024 10:53:30] "GET / HTTP/1.1" 200 -
Response received: {'response': 'I am working'}
INFO:werkzeug:127.0.0.1 - - [09/Aug/2024 10:53:36] "POST /get_quote HTTP/1.1" 200 -

quote.py

import os

from uagents import Agent, Context, Model


class Quote(Model):
    feeling: str  # Represents the user's feeling or problem
    context : str

class Response(Model):
    response: str  # Represents the response from the coach
quote = Agent(
    name="quote",
    port=8001,
    seed="QuoteAgent - Team Clutch",
    endpoint=["http://127.0.0.1:8001/submit"]
)

# Print the agent's address for reference
print(quote.address)

@quote.on_query(model=Quote)
async def handle_message(ctx: Context, sender: str, msg: Quote):
    # Define the prompt template without context
    ctx.logger.info(f"Received message from {sender}: {msg}")
    
    coach_response = "I am working"
    ctx.logger.info(f"Coach response: {coach_response}")
    
    try:
        # Send the coach's response back to the sender
        await ctx.send(sender, Response(response=coach_response))
    except Exception as e:
        ctx.logger.error(f"Failed to send response: {e}")


# Run the agent
if __name__ == "__main__":
    quote.run()

Here are the logs:

agent1q249x2snexu4f6am4rnk4e8dmuw9sqj7x886e7ydp35rnuqqgf9pu5p7yev
INFO:   [quote]: Registering on almanac contract...
INFO:   [quote]: Registering on almanac contract...complete
INFO:   [quote]: Starting server on http://0.0.0.0:8001/ (Press CTRL+C to quit)
INFO:   [quote]: Received message from user1avnfejlyy9jqgleexcthn39cdnf3lup987gez37nxnfpmm0knxcq8j989p: feeling='Happy' context='I am absolutely happy'
INFO:   [quote]: Coach response: I am working
INFO:   [quote]: Received message from user1gmkxwj6x3e4z3gkeq7yfhtp3hg03rqu35r7ymu83y4lzca0ccfpq9rzze7: feeling='happy' context=''
INFO:   [quote]: Coach response: I am working
INFO:   [quote]: Received message from user1dpmxcxf4dr7slsmzju4k2rpaavd2nha7lrusztmevxs4ut5e3dwqay8rz6: feeling='happy' context='trial'
INFO:   [quote]: Coach response: I am working

Please reach out to us if you need further assistance!

1 reply

Data-Scientist-Sahil Aug 9, 2024
Author

Hey @kshipra-fetch ,
Thank you for your immediate reply.
We figured out the error for now, its majorly because of few dependencies being missing let alone upgraded .
Thanks again for your support from Fetch AI ❤️, we are glad working with this, we will soon start contributing to this community.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

communication between two uAgents and Flask API #494

{{title}}

Replies: 1 comment 1 reply

{{title}}

{{editor}}'s edit

{{editor}}'s edit

{{title}}

Select a reply

communication between two uAgents and Flask API #494

Data-Scientist-Sahil Aug 9, 2024

Replies: 1 comment · 1 reply

kshipra-fetch Aug 9, 2024 Collaborator

Data-Scientist-Sahil Aug 9, 2024 Author

Data-Scientist-Sahil
Aug 9, 2024

Replies: 1 comment 1 reply

kshipra-fetch
Aug 9, 2024
Collaborator

Data-Scientist-Sahil Aug 9, 2024
Author