Add AWS Lambda deployment example

docs/aws-lambda.md

@@ -0,0 +1,63 @@
+# Deploying a Workflow to AWS Lambda
+
+This guide shows how to package a simple NodeTool workflow and run it as an AWS Lambda function.
+
+## Lambda Handler Example
+
+Create a handler that executes your workflow. Save the following as `examples/aws_lambda_handler.py`:
+
+```python
+#!/usr/bin/env python3
+import asyncio
+import json
+from typing import Any, Dict
+
+from nodetool.dsl.graph import graph, run_graph
+from nodetool.dsl.providers.openai import ChatCompletion
+from nodetool.metadata.types import OpenAIModel
+
+async def run_workflow(prompt: str) -> str:
+    workflow = ChatCompletion(
+        model=OpenAIModel(model="gpt-4o"),
+        messages=[{"role": "user", "content": prompt}],
+    )
+    return await run_graph(graph(workflow))
+
+
+def lambda_handler(event: Dict[str, Any], context: Any) -> Dict[str, Any]:
+    prompt = event.get("prompt", "Hello from NodeTool!")
+    result = asyncio.run(run_workflow(prompt))
+    return {"statusCode": 200, "body": json.dumps({"result": result})}
+```
+
+## Packaging and Deployment
+
+1. Install NodeTool Core and its dependencies into a local directory:
+
+```bash
+pip install nodetool-core -t ./package
+```
+
+2. Create the deployment package:
+
+```bash
+cd package
+zip -r9 ../function.zip .
+cd ..
+zip -g function.zip examples/aws_lambda_handler.py
+```
+
+3. Deploy using the AWS CLI:
+
+```bash
+aws lambda create-function \
+    --function-name nodetool-example \
+    --runtime python3.11 \
+    --handler aws_lambda_handler.lambda_handler \
+    --zip-file fileb://function.zip \
+    --role arn:aws:iam::123456789012:role/lambda-execution-role
+```
+
+Replace the role ARN with your own IAM role that has permission to run Lambda functions.
+
+Once deployed, invoke the function by sending a JSON payload with a `prompt` field.

docs/index.md

@@ -40,6 +40,7 @@ The documentation is organized into the following sections:
 - [**Agents**](agents.md) - Multi-step agent framework
 - [**Chat Module**](chat.md) - Conversational interface
 - [**Chat Providers**](chat-providers.md) - Supported LLM backends
+- [**AWS Lambda Example**](aws-lambda.md) - Deploy workflows to Lambda
 - [**Examples**](../examples/README.md) - Example workflows
 
 ## Community

examples/README.md

@@ -79,6 +79,13 @@ Before running these examples, you'll need to:
 - Integrates OpenAI's capabilities with web search
 - Shows combined AI and search functionality
 
+### 5. Deployment
+
+#### AWS Lambda Handler (`aws_lambda_handler.py`)
+
+- Demonstrates deploying a workflow as an AWS Lambda function
+- Provides a simple `lambda_handler` entry point
+
 ## Running the Examples
 
 To run any of these examples:

examples/aws_lambda_handler.py

@@ -0,0 +1,36 @@
+#!/usr/bin/env python3
+"""AWS Lambda handler example for running a NodeTool workflow.
+
+This script defines a minimal `lambda_handler` function that executes a simple
+ChatCompletion workflow using NodeTool Core. You can package this file with its
+dependencies and deploy it to AWS Lambda.
+"""
+
+import asyncio
+import json
+from typing import Any, Dict
+
+from nodetool.dsl.graph import graph, run_graph
+from nodetool.dsl.providers.openai import ChatCompletion
+from nodetool.metadata.types import OpenAIModel
+
+
+async def run_workflow(prompt: str) -> str:
+    """Execute the workflow using the provided prompt."""
+    workflow = ChatCompletion(
+        model=OpenAIModel(model="gpt-4o"),
+        messages=[{"role": "user", "content": prompt}],
+    )
+    return await run_graph(graph(workflow))
+
+
+def lambda_handler(event: Dict[str, Any], context: Any) -> Dict[str, Any]:
+    """Entry point for AWS Lambda."""
+    prompt = event.get("prompt", "Hello from NodeTool!")
+    result = asyncio.run(run_workflow(prompt))
+    return {"statusCode": 200, "body": json.dumps({"result": result})}
+
+
+if __name__ == "__main__":
+    # Simple local test
+    print(lambda_handler({"prompt": "Test"}, None))