Create README.md

Author: mwhalin

samples/agent-catalog/msft-agent-samples/foundry-agent-service-sdk/travel-planner/README.md

@@ -0,0 +1,113 @@
+# Travel Assistant Agent
+
+## Summary
+The **Travel Assistant Agent** is an AI-powered agent built using Azure AI Agent Service. It helps users receive up-to-date travel recommendations by leveraging both the **Bing Grounding Tool** and the **TripAdvisor API**. The agent summarizes relevant information clearly and offers to create a custom itinerary based on the user's travel duration.
+
+## Use Cases
+1. **Vacation Planning**: Travelers can get up-to-date suggestions on destinations, hotels, and activities.
+2. **Business Travel**: Professionals receive efficient recommendations tailored to time-sensitive trips.
+3. **Local Exploration**: Residents or tourists can find trending events, attractions, or day-trip ideas in their vicinity.
+
+## Architecture Overview
+The system consists of:
+- An AI Agent created with Azure AI Agent Service using `gpt-4o` as the base model.
+- A Bing Grounding Tool integrated via Azure Bing Account and connected to the agent.
+- A TripAdvisor API tool integrated via OpenAPI.
+- A Bicep template to automate provisioning of Azure resources.
+
+```text
++----------------+                         
+|   User Query   |                         
+| (Travel Only)  |                         
++-------+--------+                         
+        |                                  
+        v                                  
++-------------------+        invokes        +----------------------------+
+|  Travel Assistant | ------------------->  | Bing Grounding Tool (API)  |
+|     (AI Agent)    | ------------------->  | TripAdvisor API (OpenAPI)  |
++-------------------+        results        +----------------------------+
+        |                                  
+        v                                  
++------------------------------+           
+| Agent Response with summary  |           
+| and optional itinerary prompt|           
++------------------------------+           
+```
+
+## Setup Instructions
+
+### Prerequisites
+- Python 3.8+
+- Azure CLI
+- Azure subscription
+- Agent setup: deploy the latest agent setup using ([this custom deployment](https://github.com/azure-ai-foundry/foundry-samples/tree/main/use-cases/agents/setup/basic-setup)).
+  - The above creates:
+    1. AI Services resource
+    2. AI Project
+    3. Model deployment 
+- Bing Grounding resource
+- TripAdvisor API connection (via OpenAPI)
+
+### Steps
+1. **Clone the Repository**
+
+2. **Set Environment Variables**
+```bash
+PROJECT_ENDPOINT="<your-project-endpoint>" # (https://<your-ai-services-account-name>.services.ai.azure.com/api/projects/<your-project-name>)
+MODEL_DEPLOYMENT_NAME="<your-model-deployment-name>"
+BING_CONNECTION_ID="<your-bing-connection-id>"
+TRIPADVISOR_CONNECTION_ID="<your-tripadvisor-connection-id>"
+```
+
+3. **Deploy Resources Using Bicep**
+```bash
+ az deployment group create    --resource-group <your-rg>    --template-file bing-grounding.bicep    --parameters      bingAccountName="bing-grounding-agent"      bingSku="S1"      connectionName="bing-grounding-conn"      amlWorkspaceResourceName="<AI-Project-Name>"
+```
+
+4. **Run the Agent Script**
+```bash
+ python agent.py
+```
+
+## ⚙️ Configuration Guide
+| Parameter Name             | Description                                                       |
+|----------------------------|-------------------------------------------------------------------|
+| `bingAccountName`          | Unique name for the Bing grounding resource                       |
+| `bingSku`                  | Pricing tier (e.g., `S1`)                                         |
+| `connectionName`           | Name for the AML connection to Bing                               |
+| `bingTargetEndpoint`       | Bing API endpoint (defaults to `https://api.bing.microsoft.com/`) |
+| `tripadvisor_connection_id`| Azure connection ID for the TripAdvisor OpenAPI tool key          |
+| `isSharedToAll`            | Whether the connection is shared with all users                   |
+
+## Sample Data Instructions
+This repo contains a `sample_data/` directory with:
+- Sample user travel queries
+- Example travel agent responses (mocked for offline testing)
+
+You can test the interaction by invoking the agent with prompts like:
+
+## Example Agent Interaction
+
+#### 🧑 User:
+Hey, I’m visiting Mountain View this weekend with my partner. We like nature, good coffee, and local food. Can you suggest things to do?
+
+#### 🤖 Travel Agent:
+Sure! Based on Bing and Tripadvisor, here are a few local highlights:
+- **Shoreline Park**: Great for kayaking, biking, and birdwatching.
+- **Red Rock Coffee** and **Verve Coffee Roasters** are top-rated cafés downtown.
+- For food, **Eureka!** (gourmet burgers) and **Steins Beer Garden** (modern American) are local favorites.
+
+Would you like me to find availability for kayaking at Shoreline?
+
+#### 🧑 User:
+Kayaking sounds fun! Can you check if it's available Saturday morning?
+
+#### 🤖 Travel Agent:
+Yes — according to Shoreline Lake's site via Bing, kayaking is available starting 10 AM on Saturday. 
+
+## Customization Tips
+- Modify the system instructions in `template.py` to enable the agent to provide the best responses as needed.
+- Extend the agent with other useful tools or APIs (using OpenAPI spec) such as getting live flight details.
+
+## License
+This project is licensed under the MIT License. See the [LICENSE](./LICENSE) file for details.