# Sample Strands Agents Chat > A full-stack AI chat application powered by Amazon Bedrock and Strands Agents - **Type:** MCP server - **Install:** `agentstack add mcp-aws-samples-sample-strands-agents-chat` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [aws-samples](https://agentstack.voostack.com/s/aws-samples) - **Installs:** 0 - **Category:** [Cloud & Infrastructure](https://agentstack.voostack.com/c/cloud-infrastructure) - **Latest version:** 0.1.0 - **License:** MIT-0 - **Upstream author:** [aws-samples](https://github.com/aws-samples) - **Source:** https://github.com/aws-samples/sample-strands-agents-chat ## Install ```sh agentstack add mcp-aws-samples-sample-strands-agents-chat ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # 🤖 Strands Agents Chat *A full-stack AI chat application powered by Amazon Bedrock, Strands Agents, and Bedrock AgentCore* [](https://aws.amazon.com/) [](https://python.org/) [](https://reactjs.org/) [](https://typescriptlang.org/) --- ## 📸 Screenshots --- ## ✨ Features This is a sample application that serves as a base implementation for chat applications utilizing **Strands Agents**. 🏗️ **Simple Architecture** - Clean, maintainable codebase with serverless scalability 🎨 **Rich UI/UX** - Polished chat interface with intuitive user experience 🔧 **Extensible with Strands Agents** - Ready for customization with MCP, A2A, Multi Agents, Bedrock AgentCore, and more 🖼️ **Gallery Mode** - Visual gallery for viewing and managing uploaded images and files 📱 **Responsive Design** - Optimized for both desktop and mobile usage > **💡 Perfect Starting Point**: If you want to build a governance-enabled, self-hosted, high-functionality generative AI chat application on AWS, using this application as your base implementation is an excellent starting point. --- ## 🏛️ Architecture --- ## 🛠️ Built-in Tools | Tool | Description | Technology | |------|-------------|------------| | 💬 **Multi-modal Chat** | Support for images, videos, and documents | Amazon Bedrock | | 🧠 **Deep Reasoning** | Advanced AI reasoning for complex problem solving | Amazon Bedrock | | 🎨 **Image Generation** | AI-powered image creation | Nova Canvas MCP | | 📚 **AWS Documentation Search** | Search and access AWS documentation | AWS Documentation MCP | | 🔍 **Web Search** | Real-time web search capabilities | Tavily API | | 💻 **CodeInterpreter** | Execute and analyze code in real-time | Bedrock AgentCore | | 🌐 **Web Browser** | Browse web pages and analyze content | Bedrock AgentCore | --- ## 🚀 Deployment Guide ### Step 1: (Optional) Create Tavily API Key Secret Click to expand Tavily API setup instructions 1. Open [AWS Secrets Manager](https://console.aws.amazon.com/secretsmanager) 2. Click **"Store a new secret"** 3. Select **"Other type of secret"** 4. Choose **"Plaintext"** and paste your Tavily API Key 5. Create the secret and copy the **Secret ARN** ### Step 2: (Required) Configure Parameters ```bash # Navigate to CDK directory cd cdk # Install dependencies npm ci # Copy the parameter template cp parameter.template.ts parameter.ts # Edit parameter.ts with your configuration # Note: If tavilyApiKeySecretArn is null, web search tool will be disabled ``` ### Step 3: (Required) Deploy with CDK ```bash # Bootstrap CDK (run once per AWS account/region) npx cdk bootstrap # Deploy all stacks npx cdk deploy --all --require-approval never ``` 🎉 **Access your application** using the `WebUrl` from the deployment output! --- ## 💻 Development ### Frontend Development For frontend development, you can run the development server locally while connecting to your deployed backend: ```bash # Navigate to web directory cd web # Install dependencies npm ci # Start development server npm run dev ``` The development server will: - 🚀 Start at `http://localhost:5173` - 🔗 Automatically import necessary values from the `StrandsChat` stack output - 🔄 Enable hot reload for rapid development ### Pre-commit Checks Before committing your changes, ensure code quality by running the pre-check script: ```bash # Run pre-commit checks ./pre_check.sh ``` This script will validate your code formatting, run tests, and ensure everything is ready for commit. --- ## ⚙️ Customization ### 🛡️ WAF Configuration (Access Restrictions) The WAF settings are defined in `cdk/lib/waf-stack.ts`. Customize this file to modify security rules. > **Default behavior**: Allows access from all IP addresses and countries **IP Restriction Example:** ```typescript // Modify IP addresses in waf-stack.ts (line 15) const allIpv4 = new wafv2.CfnIPSet(this, 'AllowedIpv4Set', { name: 'StrandsChatIpv4', scope: 'CLOUDFRONT', ipAddressVersion: 'IPV4', addresses: ['192.168.1.0/24', '10.0.0.0/8'], // Replace with your allowed IP ranges }); ``` **GEO Restriction Example:** ```typescript // Add country-based rule to the rules array in waf-stack.ts (line 34) { name: 'GeoRestriction', priority: 1, action: { allow: {} }, statement: { geoMatchStatement: { countryCodes: ['US', 'JP', 'CA'], // Allow only these countries }, }, visibilityConfig: { cloudWatchMetricsEnabled: true, metricName: 'GeoRestriction', sampledRequestsEnabled: true, }, }, ``` > **Country Codes**: Use ISO 3166-1 alpha-2 country codes (e.g., 'US' for United States, 'JP' for Japan) ### 🔐 Disable Cognito Signup If you want to disable the Cognito signup functionality and restrict user registration: **Backend Configuration:** 1. Open `cdk/lib/strands-chat-stack.ts` 2. Find the UserPool configuration 3. Change `selfSignUpEnabled` from `true` to `false` ```typescript // In strands-chat-stack.ts (line 66-78) const userPool = new UserPool(this, 'UserPool', { selfSignUpEnabled: false, // Change from true to false signInAliases: { username: false, email: true, }, passwordPolicy: { requireUppercase: true, requireSymbols: true, requireDigits: true, minLength: 8, }, }); ``` **Frontend Configuration:** 1. Open `web/src/components/AuthWithUserPool.tsx` 2. Add the `hideSignup={true}` option to the `` component ```tsx // In AuthWithUserPool.tsx (line 75) Strands Chat ); }, }}> ``` > **Note**: After making these changes, redeploy the CDK stack for backend changes to take effect. --- ## 🏗️ Technology Stack | Layer | Technology | Purpose | |-------|------------|---------| | **Frontend** | React 19 + TypeScript + Tailwind CSS | Modern, responsive UI | | **Backend** | Python 3.13 + FastAPI + Strands Agents | AI-powered API | | **Infrastructure** | AWS CDK + Lambda + DynamoDB + S3 | Serverless architecture | | **AI/ML** | Amazon Bedrock + Claude | Advanced language models | | **Authentication** | AWS Cognito | Secure user management | --- ## 📁 Project Structure ``` ├── 🐍 api/ # Python FastAPI backend ├── ☁️ cdk/ # AWS CDK infrastructure ├── ⚛️ web/ # React frontend application └── 📖 README.md # Project documentation ``` ## Security See [CONTRIBUTING](/CONTRIBUTING.md#security-issue-notifications) for more information. ## License This library is licensed under the MIT-0 License. See the LICENSE file. ## Source & license This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [aws-samples](https://github.com/aws-samples) - **Source:** [aws-samples/sample-strands-agents-chat](https://github.com/aws-samples/sample-strands-agents-chat) - **License:** MIT-0 Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/mcp-aws-samples-sample-strands-agents-chat - Seller: https://agentstack.voostack.com/s/aws-samples - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.