{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_mg3kaynd7wgg","handle":"iflow-mcp-kyopark2014-agent-runtime-use-aws","url":"https://wellknown.network/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws","links":{"self":"https://wellknown.network/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws/record.json","html":"https://wellknown.network/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws","markdown":"https://wellknown.network/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws/record.md","api":"https://wellknown.network/api/v1/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws","status":"https://wellknown.network/api/v1/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws/status","claim":"https://wellknown.network/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws/claim.json","badge":"https://wellknown.network/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws/history","tools":"https://wellknown.network/api/v1/agents/iflow-mcp-kyopark2014-agent-runtime-use-aws/tools"},"ard":{"identifier":"urn:air::server:iflow-mcp-kyopark2014-agent-runtime-use-aws","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"iflow-mcp_kyopark2014-agent-runtime-use-aws","summary":"MCP server for AWS service operations using boto3","description":"# Agent와 MCP 서버의 배포 및 활용\n\n여기에서는 AgentCore Runtime을 이용해서 1) LangGraph, Strands SDK, Claude Agent SDK를 이용해 만든 agent를 배포하는 방법과 2) agent에 필요한 데이터를 수집하고 사용자의 의도에 따른 동작을 수행하는 방법을 설명합니다. AgentCore는 Agent와 MCP를 위한 서버리스 production 환경으로서 Agent와 MCP 서버를 편리하게 배포하고 안전하고 효과적으로 운용할 수 있습니다.\n\n## 주요 구현 \n\n### 전체 Architecture\n\n전체적인 Architecture는 아래와 같습니다. 여기서는 MCP를 지원하는 Strands와 LangGraph agent를 [AgentCore](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/what-is-bedrock-agentcore.html)를 이용해 배포하고 streamlit 애플리케이션을 이용해 사용합니다. 개발자는 각 agent에 맞는 [Dockerfile](./runtime/langgraph/Dockerfile)을 이용하여, docker image를 생성하고 ECR에 업로드 합니다. 이후 [bedrock-agentcore-control](https://docs.aws.amazon.com/bedrock-agentcore-control/latest/APIReference/Welcome.html)의 [create_agent_runtime.py](./runtime/langgraph/create_agent_runtime.py)을 이용해서 [AgentCore](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/what-is-bedrock-agentcore.html)의 runtime으로 배포합니다. 이 작업이 끝나면 EC2와 같은 compute에 있는 streamlit에서 LangGraph와 Strands agent를 활용할 수 있습니다. 애플리케이션에서 AgentCore의 runtime을 호출할 때에는 [bedrock-agentcore](https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/bedrock-agentcore.html)의 [invoke_agent_runtime](https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/bedrock-agentcore/client/invoke_agent_runtime.html)을 이용합니다. 이때에 각 agent를 생성할 때에 확인할 수 있는 [agentRuntimeArn](https://docs.aws.amazon.com/bedrock-agentcore-control/latest/APIReference/API_Agent.html)을 이용합니다. Agent는 [MCP](https://modelcontextprotocol.io/introduction)을 이용해 RAG, AWS Document, Tavily와 같은 검색 서비스를 활용할 수 있습니다. 여기에서는 RAG를 위하여 Lambda를 이용합니다. 데이터 저장소의 관리는 Knowledge base를 사용하고, 벡터 스토어로는 OpenSearch를 이용합니다. Agent에 필요한 S3, CloudFront, OpenSearch, Lambda등의 배포를 위해서는 AWS CDK를 이용합니다.\n\n<img width=\"850\" alt=\"image\" src=\"https://github.com/user-attachments/assets/efce9789-6287-4b49-8c43-3485d1a9fd35\" />\n\nAgentCore의 runtime은 배포를 위해 Docker를 이용합니다. 현재(2025.7) 기준으로 arm64와 1GB 이하의 docker image를 지원합니다.\n\n### AgentCore 소개\n…","publisher":null,"homepage":null,"repository":null,"version":"0.1.0","license":null,"protocols":["mcp"],"tags":["mcp"],"pricing":null,"endpoints":[{"url":"pypi:iflow-mcp_kyopark2014-agent-runtime-use-aws","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","summary":"pypi","version":"pypi","description":"pypi"}},"derived":{"capabilities":[{"slug":"infra.cloud","name":"Cloud Platforms","confidence":1,"provenance":"derived"},{"slug":"dev.docs-lookup","name":"Documentation Lookup","confidence":0.791,"provenance":"derived"}],"categories":["dev","infra"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"iflow-mcp_kyopark2014-agent-runtime-use-aws","registry":"pypi","observedAt":"2026-09-15T18:22:09.083Z","publishedAt":"2026-02-11T11:49:55.332396Z","latestVersion":"0.1.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"iflow-mcp_kyopark2014-agent-runtime-use-aws","url":"https://pypi.org/project/iflow-mcp_kyopark2014-agent-runtime-use-aws/","firstSeenAt":"2026-09-09T19:24:44.241Z","fetchedAt":"2026-09-15T18:21:12.116Z","normalizedAt":"2026-09-15T18:21:12.116Z"}]},"firstSeenAt":"2026-09-09T19:24:44.241Z","updatedAt":"2026-09-15T18:22:09.083Z"}