aws-samples/bedrock-engineer
Universal AI Agent using Amazon Bedrock, capable of customize to create/edit files, execute commands, search the web, use knowledge base, use multi-agents, generative images and more.
What it solves
Bedrock Engineer is an autonomous software development agent application that streamlines the coding process. It removes the dependency on specific IDEs (like VS Code) by providing its own dedicated interface for AI-driven development, allowing for richer interactive experiences and visual diagramming.
How it works
Built on Amazon Bedrock, the application allows users to deploy autonomous agents that can interact with the local file system, execute system commands, and perform web searches. Users can customize agents by defining their system prompts and selecting specific tools from a library of capabilities, including file operations, web search, and Bedrock-specific integrations (like image and video generation). It also supports the Model Context Protocol (MCP) to connect to external tool servers.
Who it’s for
It is designed for software developers, programming students, and product designers who want an AI-powered assistant to handle code generation, project structure management, and data analysis without being tied to a specific editor.
Highlights
- Autonomous Development: Capable of creating/editing files, executing commands, and managing project structures.
- Customizable Agents: Users can create specialized agents (e.g., Software Developer, Programming Mentor) with custom system prompts.
- Tool Integration: Includes a wide array of tools for file system access, web search (Tavily), and Amazon Bedrock services (Nova Reel, Knowledge Bases, and Flows).
- Background Automation: Supports scheduled task execution using cron expressions for continuous workflows.
- Agent Directory: A community hub for discovering and sharing pre-configured agents.
- Secure Execution: Features a Python code interpreter running in a secure Docker environment.