Back to Prompt Library
planning
OpenAI Agents SDK Initialization and Agent Roles
Inspect the original prompt language first, then copy or adapt it once you know how it fits your workflow.
Linked challenge: Strategic Market Expansion Agent Team with OpenAI Agents SDK and Multi-Model Inference
Format
Code-aware
Lines
14
Sections
1
Linked challenge
Strategic Market Expansion Agent Team with OpenAI Agents SDK and Multi-Model Inference
Prompt source
Original prompt text with formatting preserved for inspection.
14 lines
1 sections
No variables
1 code block
Initialize an OpenAI Assistant with the 'GPT-5 Pro-mini' model to act as a 'Strategic Orchestrator'. Define two 'Expert Analysts' as custom tools callable by the Orchestrator: one for 'Qualitative Market Sentiment' (backed by Claude 4 Sonnet via Together AI) and another for 'Quantitative Financial Modeling' (backed by Gemini 3 Flash via Together AI). Provide Python code for the initial setup. ```python
from openai import OpenAI client = OpenAI(api_key="YOUR_OPENAI_API_KEY") # Define tools for the orchestrator to call
# These tools will internally route to Together AI for Claude/Gemini
def qualitative_analysis_tool(topic: str): # This would call Together AI with Claude 4 Sonnet print(f"Calling Claude 4 Sonnet via Together AI for qualitative analysis on: {topic}") return "Qualitative insights from Claude 4 Sonnet..." def quantitative_modeling_tool(data: str): # This would call Together AI with Gemini 3 Flash print(f"Calling Gemini 3 Flash via Together AI for quantitative modeling with: {data}") return "Quantitative projections from Gemini 3 Flash..." # Example of creating an Assistant (Orchestrator)
# assistant = client.beta.assistants.create(
# name="Strategic Orchestrator",
# instructions="You are an expert strategic analyst. Use the provided tools to gather qualitative and quantitative insights...",
# model="GPT-5 Pro-mini",
# tools=[
# {"type": "function", "function": {"name": "qualitative_analysis_tool", "parameters": {...}}},
# {"type": "function", "function": {"name": "quantitative_modeling_tool", "parameters": {...}}}
# ]
# )
```Adaptation plan
Keep the source stable, then change the prompt in a predictable order so the next run is easier to evaluate.
Keep stable
Preserve the role framing, objective, and reporting structure so comparison runs stay coherent.
Tune next
Swap in your own domain constraints, anomaly thresholds, and examples before you branch variants.
Verify after
Check whether the prompt asks for the right evidence, confidence signal, and escalation path.