Neha Thakur
ff11ccbb89
* feat(video-games-sales-assistant): Add all AgentCore features - Gateway, Policy Engine, Guardrails, Evaluators, Observability - Add AgentCore Gateway with MCP Lambda target for PostgreSQL tools - Add Cedar Policy Engine (allow default, block PII/cost columns) - Add Bedrock Guardrail for content filtering (PII + cost topics) - Add LLM-as-a-Judge evaluators (SqlAccuracy, ResponseQuality) - Add Gateway observability (CloudWatch Logs delivery) - Add deploy.py for script-based deployment alternative - Add evaluations/evaluate.py harness with 8 test scenarios - Add setup-frontend.sh for automated frontend configuration - Expand README to 484 lines covering all AgentCore features * fix(amplify): Add monorepo amplify.yml at repo root for Amplify Hosting * fix(amplify): Add esbuild as direct devDependency for Amplify Hosting builds pnpm exec cannot find esbuild binary when it is only a transitive dependency. Adding it directly ensures the binary is linked in .bin/. Also removed --frozen-lockfile to allow platform-specific resolution. * fix(amplify): Fix Amplify Hosting build - update lockfile, skip backend deploy * fix(amplify): Generate amplify_outputs.json at build time from env vars * fix(amplify): Generate amplify_outputs.json at build time from env vars * docs: Update READMEs for Amplify Hosting, remove amplify_outputs.json from tracking * docs: Rewrote parent README as self-sufficient deploy guide, added architecture diagram * Add nehatb to CONTRIBUTORS.md * feat: Replace standalone Bedrock Guardrail with AgentCore guardrails-in-policy - Use Cedar guardrails-in-policy at Gateway Policy Engine layer - Move SQL content validation to Lambda handler - Remove standalone AWS::Bedrock::Guardrail and GUARDRAIL_ID env vars - Remove deploy-policies.sh (policies now in single CDK deploy) - Rename video-games-sales-assistant to data-analyst-conversational-assistant - Update architecture diagram * fix: Address PR review feedback - batch evals, guardrail docs, agentcore CLI reference - Replace manual scenario iteration with BatchEvaluationRunner using native Dataset management API for server-side batch evaluation - Add guardrail testing section with curl examples for prompt injection, harmful content, and PII suppression - Remove agentcore-cli tag from agentcore.json * fix: ruff formatting and missing time import in data-analyst-conversational-assistant --------- Signed-off-by: Bharathi Srinivasan <bhrsrini@amazon.com> Co-authored-by: nehatb <nehatb@amazon.com> Co-authored-by: Bharathi Srinivasan <bhrsrini@amazon.com>
Amazon Bedrock AgentCore Use Cases
End-to-end samples organized by agent type. Each folder maps to one of the three workload categories used in AgentCore documentation.
Categories
01-conversational-agents
- AWS Operations Agent: Intelligent AWS operations assistant with Okta authentication and comprehensive monitoring capabilities
- Customer Support Assistant: Production-ready customer service agent with memory, knowledge base integration, and Google OAuth
- DB Performance Analyzer: Database performance monitoring and analysis agent with PostgreSQL integration
- Device Management Agent: IoT device management system with Cognito authentication and real-time monitoring
- Enterprise Web Intelligence Agent: Web research and analysis agent using browser tools for competitive intelligence
- Farm Management Advisor: Agricultural advisory system with plant detection, weather forecasting, and care recommendations
- Auth0 Multi-Agent OBO: RFC 8693 On-Behalf-Of token exchange with Auth0 PKCE — coordinator exchanges user JWT for attenuated per-agent tokens in a multi-agent financial services system
- Finance Personal Assistant: Personal budget management with multi-agent workflows and guardrails
- Healthcare Appointment Agent: FHIR-compliant healthcare appointment scheduling with patient data integration
- Local Prototype to AgentCore: Migration guide from local development to production AgentCore deployment
- Market Trends Agent: Financial market analysis with browser tools and memory integration
- SRE Agent: Site reliability engineering assistant with multi-agent LangGraph workflows
- Text to Python IDE: Code generation and execution environment with AgentCore Code Interpreter
- Data Analyst Conversational Assistant: Data analysis assistant with Amplify frontend and CDK deployment
- Claude Code Gateway MCP Server: Integrate Claude Code with MCP Server using AgentCore Gateway for dynamic tool loading and centralized access Agents that interact with users in real time. Users authenticate through an identity provider, the agent maintains session and long-term memory per user, and responses stream back as the agent works. See the category README for the full list and a guide on which sample to start with.
| Sample | Vertical | Key Features |
|---|---|---|
| A2A-multi-agent-incident-response | IT / DevOps | Runtime, Gateway, Memory, A2A (3 frameworks) |
| AWS-operations-agent | Cloud Operations | Runtime, Gateway, Memory, Policy, Observability |
| customer-support-assistant-vpc | Retail / E-commerce | Runtime, Gateway (VPC) |
| deep-research-agent | Research / Q&A | Gateway (Web Search), Runtime |
| device-management-agent | IoT / Smart Home | Runtime, Gateway, Policy, Identity (Cognito) |
| finance-personal-assistant | Personal Finance | Gateway, Policy |
| healthcare-appointment-agent | Healthcare | Runtime, Gateway, Policy, Observability (FHIR R4) |
| lakehouse-agent | Data and Analytics | Runtime, Gateway, Memory, Policy (row-level security) |
| market-trends-agent | Financial Services | Runtime, Memory, Browser, Evaluations, Optimization |
| SRE-agent | Site Reliability | Runtime, Gateway, Memory, Observability |
| video-games-sales-assistant | Retail / Gaming | Runtime, Gateway, Memory |
02-workflow-automation-agents
Agents that run without a user in the loop. They are triggered by events such as file uploads, webhook calls, or scheduled jobs. Identity is service-to-service rather than user-facing, and memory is minimal since state is carried in the event payload.
| Sample | Vertical | Key Features |
|---|---|---|
| event-driven-claims-agent | Insurance | Runtime, Gateway, Memory, Policy, Evaluations, Observability |
| visa-b2b-account-payable-agent | B2B Payments | Runtime, Gateway, Policy, Payments |
| enterprise-web-intelligence-agent | Market Intelligence | Runtime, Browser |
| intelligent-event-agent | General / Events | Runtime, Memory, Gateway (in development) |
| multi-isv-orchestration | Enterprise CRM + ERP | Gateway (multi-target), Identity (Cognito + CustomOauth2) |
03-coding-assistants
Agents that help developers write, run, or fix code. Tasks tend to be longer-running and scoped to a project or repository. AgentCore Code Interpreter handles sandboxed execution, and Gateway can aggregate multiple developer tool APIs behind one MCP endpoint.
| Sample | Use Case | Key Features |
|---|---|---|
| text-to-python-ide | Text-to-Python IDE with sandboxed execution | Runtime, Code Interpreter, Memory, Policy |
| claude-code-gateway-mcp-server | Single MCP endpoint for Claude Code | Gateway, Identity |