[{"data":1,"prerenderedAt":91},["ShallowReactive",2],{"reinvent2026-svs324":3},{"id":4,"title":5,"YouTubeChannelID":6,"banner":6,"contributors":7,"description":6,"extension":10,"introBox":11,"meta":14,"pdf":15,"resources":18,"stem":89,"type":6,"video":6,"videos":6,"__hash__":90},"reinvent2026\u002Freinvent2026\u002Fsvs324.json","SVS324: Building agentic apps with serverless: Lambda as your agent's toolbox",null,[8,9],"content\u002Fcontributors\u002Furi-segev.json","content\u002Fcontributors\u002Fandre-stoll.json","json",{"headline":12,"text":13},"Resources for SVS324: Building agentic apps with serverless: Lambda as your agent's toolbox","Agents need tools to retrieve data, take actions and reach external systems, and designing AWS Lambda functions as agent tools takes different thinking than building API backends. This chalk talk worked through a multi-tenant commerce platform with two patterns: exposing existing Lambda functions as MCP tools through Amazon Bedrock AgentCore Gateway, and running an MCP server on AWS Lambda Web Functions. It covered tool design and identity propagation from the caller's token to the tool.\u003Cbr>\u003Cbr>This page has the slides and the links for each part of the talk.",{},{"source":16,"headline":17},"https:\u002F\u002Fserverlessland.s3.us-east-1.amazonaws.com\u002Fpdf\u002Freinvent2026\u002FSVS324.pdf","Slides from the presentation",{"left":19,"right":51},[20,34],{"headline":21,"subheading":22,"text":23,"bullets":24},"MCP for agent tools","The protocol","MCP is how remote tools reach agents. A tool holds the business logic, and one MCP server can host many tools.",[25,28,31],{"text":26,"link":27},"What is the Model Context Protocol?","https:\u002F\u002Fmodelcontextprotocol.io\u002Fdocs\u002Fgetting-started\u002Fintro",{"text":29,"link":30},"Understanding MCP servers","https:\u002F\u002Fmodelcontextprotocol.io\u002Fdocs\u002Flearn\u002Fserver-concepts",{"text":32,"link":33},"MCP specification","https:\u002F\u002Fmodelcontextprotocol.io\u002Fspecification\u002Flatest",{"headline":35,"subheading":36,"text":37,"bullets":38},"AgentCore Gateway: Lambda functions and MCP servers as targets","Expose existing functions as tools","Put Amazon Bedrock AgentCore Gateway in front of the Lambda functions that already back your APIs, or in front of your own MCP server, and the Gateway handles the MCP protocol.",[39,42,45,48],{"text":40,"link":41},"Amazon Bedrock AgentCore Gateway","https:\u002F\u002Fdocs.aws.amazon.com\u002Fbedrock-agentcore\u002Flatest\u002Fdevguide\u002Fgateway.html",{"text":43,"link":44},"AWS Lambda function targets","https:\u002F\u002Fdocs.aws.amazon.com\u002Fbedrock-agentcore\u002Flatest\u002Fdevguide\u002Fgateway-add-target-lambda.html",{"text":46,"link":47},"MCP server targets","https:\u002F\u002Fdocs.aws.amazon.com\u002Fbedrock-agentcore\u002Flatest\u002Fdevguide\u002Fgateway-target-MCPservers.html",{"text":49,"link":50},"Move AI agents from proof of concept to production with AgentCore","https:\u002F\u002Faws.amazon.com\u002Fblogs\u002Fmachine-learning\u002Fmove-your-ai-agents-from-proof-of-concept-to-production-with-amazon-bedrock-agentcore\u002F",[52,58,72],{"headline":53,"subheading":54,"text":55,"button":56},"MCP servers on AWS Lambda Web Functions","Docs","Run an MCP SDK server on Lambda Web Functions when you offer tools, resources and prompts to other people's agents, with AgentCore Gateway in front.",{"text":54,"link":57},"https:\u002F\u002Fdocs.aws.amazon.com\u002Flambda\u002Flatest\u002Fdg\u002Fweb-functions.html",{"headline":59,"subheading":60,"text":61,"bullets":62},"Well-Architected Agentic AI Lens","Designing the tools","Best practices for tool design: strict schemas, validated input and output, and meta-tools that cut reasoning loops.",[63,66,69],{"text":64,"link":65},"Agentic AI Lens","https:\u002F\u002Fdocs.aws.amazon.com\u002Fwellarchitected\u002Flatest\u002Fagentic-ai-lens\u002Fagentic-ai-lens.html",{"text":67,"link":68},"AGENTSEC02-BP02 Validate tool inputs and outputs","https:\u002F\u002Fdocs.aws.amazon.com\u002Fwellarchitected\u002Flatest\u002Fagentic-ai-lens\u002Fagentsec02-bp02.html",{"text":70,"link":71},"AGENTPERF06-BP03 Optimize meta-tool utilization and tool chaining","https:\u002F\u002Fdocs.aws.amazon.com\u002Fwellarchitected\u002Flatest\u002Fagentic-ai-lens\u002Fagentperf06-bp03.html",{"headline":73,"subheading":74,"text":75,"bullets":76},"Identity from token to tool","Multi-tenant isolation","The tenant comes from a validated token, never from a model argument. AgentCore Gateway validates inbound identity, applies policy and interceptors, and handles outbound credentials.",[77,80,83,86],{"text":78,"link":79},"Fine-grained access control with AgentCore Gateway interceptors","https:\u002F\u002Faws.amazon.com\u002Fblogs\u002Fmachine-learning\u002Fapply-fine-grained-access-control-with-bedrock-agentcore-gateway-interceptors\u002F",{"text":81,"link":82},"Secure AI agents with policy and Lambda interceptors in AgentCore Gateway","https:\u002F\u002Faws.amazon.com\u002Fblogs\u002Fmachine-learning\u002Fsecure-ai-agents-with-policy-and-lambda-interceptors-in-amazon-bedrock-agentcore-gateway\u002F",{"text":84,"link":85},"Pool-model multi-tenancy with Amazon Bedrock AgentCore","https:\u002F\u002Faws.amazon.com\u002Fblogs\u002Fmachine-learning\u002Fshared-infrastructure-isolated-tenants-pool-model-multi-tenancy-with-amazon-bedrock-agentcore\u002F",{"text":87,"link":88},"Set up outbound authorization for your gateway","https:\u002F\u002Fdocs.aws.amazon.com\u002Fbedrock-agentcore\u002Flatest\u002Fdevguide\u002Fgateway-outbound-auth.html","reinvent2026\u002Fsvs324","7ppBRZqoGBua1Z5_o8xovl8piE3hzfiBH-CEjiOUhlc",1794310836599]