Comet Opik & Open AI Warsaw Hackathon

README.md

Recipe Agent Demo with OpenAI Agents SDK + Comet Opik

A simple demo showcasing the integration of OpenAI Agents SDK with Comet Opik for observability. This demo features two agents:

1. RecipeSuggesterAgent: Suggests recipes based on provided ingredients
2. RecipeResearchAgent: Researches additional information about the suggested recipe

Setup

1. Install dependencies:

pip install openai-agents opik python-dotenv

2. Set up your OpenAI API key:

# Option A: Create a .env file
OPENAI_API_KEY=your_openai_key_here

# Option B: Export as environment variable
export OPENAI_API_KEY=your_openai_key_here

3. Configure Opik:

opik configure
# Follow prompts for local server address

4. Run the demo:

python recipe_agent.py

Features

• Multi-agent orchestration using OpenAI Agents SDK
• Asynchronous execution with proper error handling
• Integration with Comet Opik for tracing and observability
• Simple command-line interface for ingredient input

Example Usage

Welcome to the Recipe Agent Demo!

Enter a list of ingredients (comma-separated): chicken, rice, soy sauce

Suggested Recipe:
[Recipe suggestion will appear here]

Recipe Research:
[Additional information will appear here]

Project Structure

• recipe_agent.py: Main application file containing agent definitions and logic
• .env: (Optional) Environment variables file for API keys

recipe_agent.py

"""
Sample Recipe Agent Demo for OpenAI Hackathon + Comet Opik

Setup Instructions:
-------------------
1. Install dependencies:
    pip install openai-agents opik python-dotenv

2. Set up OpenAI API key:
   Option A) Create a .env file in the project root:
    OPENAI_API_KEY=your_openai_key_here
    
   Option B) Export as environment variable:
    export OPENAI_API_KEY=your_openai_key_here

3. Configure Opik (for tracing and online evals):
    opik configure
   # Follow prompts for local server address

4. Run the script:
    python recipe_agent.py

References:
- Opik integration: https://www.comet.com/docs/opik/tracing/integrations/openai_agents
- OpenAI Agents SDK: https://platform.openai.com/docs/guides/agents
"""

import os
import asyncio
from dotenv import load_dotenv
from agents import Agent, Runner, set_trace_processors
from opik.integrations.openai.agents import OpikTracingProcessor

# Load environment variables from .env file
load_dotenv()

# Verify OpenAI API key is set
if not os.getenv("OPENAI_API_KEY"):
    raise ValueError("OPENAI_API_KEY environment variable is not set")

# --- Agent 1: Recipe Suggester ---
RecipeSuggesterAgent = Agent(
    name="RecipeSuggester",
    instructions=(
        "You are a helpful chef. Given a list of ingredients, suggest a creative recipe. "
        "Be concise and clear. Output the recipe name and steps."
    ),
)

# --- Agent 2: Recipe Researcher ---
RecipeResearchAgent = Agent(
    name="RecipeResearcher",
    instructions=(
        "You are a research assistant. Given a recipe name, research and provide "
        "relevant information, tips, or variations for the recipe. Summarize your findings."
    ),
)

# --- Enable Opik Tracing ---
set_trace_processors(processors=[OpikTracingProcessor()])

# --- Demo Flow ---
async def main():
    print("Welcome to the Recipe Agent Demo!\n")
    ingredients = input("Enter a list of ingredients (comma-separated): ")

    # Step 1: Suggest a recipe
    recipe_result = await Runner.run(
        RecipeSuggesterAgent,
        f"Ingredients: {ingredients}"
    )
    recipe_name = recipe_result.final_output.split('\n')[0]  # Assume first line is recipe name

    print("\nSuggested Recipe:\n", recipe_result.final_output)

    # Step 2: Research the recipe
    research_result = await Runner.run(
        RecipeResearchAgent,
        f"Research the recipe: {recipe_name}"
    )
    print("\nRecipe Research:\n", research_result.final_output)

if __name__ == "__main__":
    asyncio.run(main())

recipe_evals.md

Recipe Agent Online Evaluation Prompts

These prompts are designed for Opik's online evaluation system to assess recipe quality and usability.

1. Recipe Taste Appeal Evaluation

You are a professional chef and food critic. Evaluate the suggested recipe's potential taste appeal and flavor combinations. Consider:
- Ingredient harmony and balance
- Seasoning and flavor depth
- Cultural authenticity (if applicable)
- Overall gastronomic appeal

Rate on a scale of 1-10 and explain your rating with specific culinary insights.

INPUT:
{{input}}

OUTPUT:
{{output}}

EVALUATION FORMAT:
Taste Score: [1-10]
Reasoning: [One clear sentence focusing on flavor profile and culinary merit]

2. Recipe Complexity and Accessibility

You are an experienced cooking instructor who specializes in home cooking. Evaluate the recipe's practicality and ease of execution. Consider:
- Ingredient accessibility
- Required cooking skills
- Time commitment
- Equipment needs
- Clear instructions

Rate on a scale of 1-10 (where 10 is extremely beginner-friendly) and provide specific implementation insights.

INPUT:
{{input}}

OUTPUT:
{{output}}

EVALUATION FORMAT:
Complexity Level: [Beginner/Intermediate/Advanced]
Time Estimate: [Quick (<30min)/Medium (30-60min)/Extended (>60min)]

Response example should be complexity + time, i.e: Intermediate 40mins

3. Overall Recipe Quality

You are an expert recipe developer and cookbook editor. Evaluate the overall quality and completeness of the recipe response. Consider:
- Ingredient specifications
- Step-by-step clarity
- Technical accuracy
- Safety considerations
- Serving size/yield information
- Additional tips/variations provided

Rate on a scale of 1-10 and explain your comprehensive assessment.

INPUT:
{{input}}

OUTPUT:
{{output}}

EVALUATION FORMAT:
Quality Score: [1-10]
Only give the quality score back

Using These Prompts

1. Add these evaluations to your Opik configuration
2. Monitor real-time recipe quality metrics
3. Use insights to improve agent instructions and outputs
4. Track quality trends over time

These prompts help ensure recipes are:

• Tasty and appealing
• Practical for home cooks
• Well-documented and complete