Use case
i'd like to be able to write serverless MCP servers with ease and integrate all my powertools capabilities:
swagger, event validation, event resolver, logger, tracing, metrics etc.
Solution/User Experience
I'd like to have UX like in https://docs.powertools.aws.dev/lambda/python/latest/core/event_handler/api_gateway/#response-auto-serialization but combined with the official python SDK for MCP server with all the right config to make it work on a lambda.
# server.py
from aws_lambda_powertools.event_handler import MCPResolver
mcp = MCPResolver('myMcp)
# Add an addition tool
@mcp.tool()
def add(a: int, b: int) -> int:
"""Add two numbers"""
return a + b
# Add a dynamic greeting resource
@mcp.resource("greeting://{name}")
def get_greeting(name: str) -> str:
"""Get a personalized greeting"""
return f"Hello, {name}!"
# You can continue to use other utilities just as before
@logger.inject_lambda_context()
@tracer.capture_lambda_handler
def lambda_handler(event: dict, context: LambdaContext) -> dict:
return mcp.resolve(event, context)Alternative solutions
use unknown open sources like MCPEngine and build my own wrappers for logging/tracer.. i want the powertools edition :)
Acknowledgment
Use case
i'd like to be able to write serverless MCP servers with ease and integrate all my powertools capabilities:
swagger, event validation, event resolver, logger, tracing, metrics etc.
Solution/User Experience
I'd like to have UX like in https://docs.powertools.aws.dev/lambda/python/latest/core/event_handler/api_gateway/#response-auto-serialization but combined with the official python SDK for MCP server with all the right config to make it work on a lambda.
Alternative solutions
Acknowledgment