Add python-story-agents: writer + reviewer multi-agent app using Microsoft Agent Framework SDK

python-story-agents/.env.example

@@ -0,0 +1,4 @@
+# Azure AI Foundry connection info
+AZURE_AI_PROJECT_ENDPOINT=https://<your-project>.services.ai.azure.com/api/projects/<your-project>
+AZURE_AI_API_KEY=<your-api-key>
+AZURE_AI_MODEL_DEPLOYMENT_NAME=<your-model-deployment-name>

python-story-agents/README.md

@@ -0,0 +1,35 @@
+# python-story-agents
+
+A simple multi-agent app using the [Microsoft Agent Framework SDK](https://github.com/microsoft/agent-framework).
+
+- **Writer Agent**: Writes a short story based on user input.
+- **Reviewer Agent**: Reviews the story and outputs a score from 0 to 100.
+
+Both agents use an Azure AI Foundry model.
+
+## Setup
+
+1. Install dependencies:
+   ```bash
+   pip install -r requirements.txt
+   ```
+
+2. Copy `.env.example` to `.env` and fill in your Azure AI Foundry connection info:
+   ```bash
+   cp .env.example .env
+   ```
+
+3. Edit `.env` with your actual values:
+   - `AZURE_AI_PROJECT_ENDPOINT` – Your Azure AI project endpoint
+   - `AZURE_AI_API_KEY` – Your Azure AI API key
+   - `AZURE_AI_MODEL_DEPLOYMENT_NAME` – Your model deployment name
+
+## Run
+
+```bash
+python main.py
+```
+
+Enter a topic or prompt when asked, and the app will:
+1. Have the **Writer Agent** write a short story
+2. Have the **Reviewer Agent** score it (0–100) with brief feedback

python-story-agents/main.py

@@ -0,0 +1,53 @@
+import asyncio
+import os
+
+from agent_framework import Agent
+from agent_framework.azure import AzureAIChatClient
+
+
+AZURE_AI_PROJECT_ENDPOINT = os.getenv("AZURE_AI_PROJECT_ENDPOINT", "<your-azure-ai-project-endpoint>")
+AZURE_AI_API_KEY = os.getenv("AZURE_AI_API_KEY", "<your-azure-ai-api-key>")
+AZURE_AI_MODEL_DEPLOYMENT_NAME = os.getenv("AZURE_AI_MODEL_DEPLOYMENT_NAME", "<your-model-deployment-name>")
+
+
+async def main():
+    user_input = input("Enter a topic or prompt for the short story: ").strip()
+    if not user_input:
+        user_input = "a brave knight and a mysterious dragon"
+
+    client = AzureAIChatClient(
+        endpoint=AZURE_AI_PROJECT_ENDPOINT,
+        api_key=AZURE_AI_API_KEY,
+        deployment_name=AZURE_AI_MODEL_DEPLOYMENT_NAME,
+    )
+
+    writer = Agent(
+        client=client,
+        name="Writer",
+        instructions=(
+            "You are a creative short story writer. "
+            "Write an engaging short story (around 150-200 words) based on the user's input."
+        ),
+    )
+
+    reviewer = Agent(
+        client=client,
+        name="Reviewer",
+        instructions=(
+            "You are a literary critic. Review the provided short story and output a score from 0 to 100 "
+            "based on creativity, coherence, and engagement. "
+            "Respond with: Score: <number>\nFeedback: <brief feedback>"
+        ),
+    )
+
+    print("\n--- Writer Agent ---")
+    story = await writer.run(user_input)
+    print(story)
+
+    print("\n--- Reviewer Agent ---")
+    review = await reviewer.run(f"Please review this story:\n\n{story}")
+    print(review)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

python-story-agents/requirements.txt

@@ -0,0 +1 @@
+agent-framework-azure-ai --pre