AWS re:Invent 2025 Day 3: AWS launched nine new agent features to help developers deploy agents faster. Key updates include TypeScript support in Strands Agents SDK for front-end engineers, ARM/x86 edge device compatibility enabling sub-100ms latency for real-time path planning agents on vehicle chips, and enhancements to Amazon Bedrock.....
AWS launches Amazon Bedrock Custom Model Import, enabling deployment of 2B and 12B parameter GPT-OSS open-source models. It simplifies migration with API compatibility, automated GPU setup, and S3-based model uploads.....
Anthropic launches Claude Opus 4.5 on Amazon Bedrock, outperforming Sonnet 4.5 and Opus 4.1 in coding, AI agents, computer operations, and office tasks at one-third the cost of Opus series.....
Amazon launches the AI Agents platform AgentCore and integrates it into Amazon Bedrock, entering the intelligent agent infrastructure field. The platform helps enterprises quickly build and deploy AI agents, supporting major models such as OpenAI and Gemini. Developers no longer need to build systems from scratch, and can simplify the integration process using Amazon's environment.
All your data, in order.
Anthropic
$21
Input tokens/M
$105
Output tokens/M
200
Context Length
A tool for querying AWS cost data and Bedrock usage logs via the MCP protocol
An SSE-based MCP server providing image and video generation tools
A tool for monitoring the status of Minecraft servers, supporting Bedrock and Java editions, providing a Web API interface and database persistence function.
An MCP server based on TypeScript, which controls the Minecraft Bedrock Edition via WebSocket connection, provides functions such as player operations, world management, and building tools, and supports AI automated operations.
A reference project implementing the Model Context Protocol (MCP), providing a standard for tool calls between LLMs and applications, including local and remote call implementations, and integrating AWS Bedrock and Claude 3.7.
An unstable but efficient Minecraft Bedrock Edition server that runs on arm64 devices (such as Raspberry Pi), based on the mcpelauncher project and supports the latest version v1.21.2.
A PowerPoint translation tool based on Amazon Bedrock, supporting command - line and FastMCP service integration, and performing high - quality translation while preserving the original format.
An MCP server based on the Nova Canvas model on Amazon Bedrock, supporting multiple image generation and editing functions.
The Agentic MCP Client is an independent AI agent runner that interacts with models such as Claude, BedRock, and OpenAI through MCP protocol tools, supporting autonomous cloud operation and secure system interaction.
An AI image generation MCP service based on the Amazon Bedrock Nova Canvas model
This project demonstrates a serverless MCP (Model Context Protocol) tool implementation based on AWS Lambda, including a Python server - side and a TypeScript client, supporting streaming HTTP communication. The server - side simplifies the MCP tool development process, has built - in session state management, and demonstrates integration with Amazon Bedrock.
This project demonstrates how to integrate the conversational AI capabilities of AWS Bedrock with relational databases through the MCP server architecture to achieve the function of querying databases in natural language.
A TypeScript - based MCP server that controls Minecraft Bedrock Edition via WebSocket connection, providing functions such as player control, world operation, and building tools, and supporting integration with MCP clients for AI automation.
A tool for querying non - connected information of Minecraft Bedrock Edition servers, which displays relevant information by entering the server address.
An MCP server based on the Amazon Bedrock Nova Canvas model, providing high-quality AI image generation services, supporting functions such as text description image generation, negative prompt optimization, size configuration, and seed control.
A command-line tool and API for analyzing AWS Bedrock usage logs through Anthropic's MCP protocol, providing functions such as model usage statistics, user analysis, and natural language query.
An MCP server project based on the Spring AI framework, using the Sync SSE transmission protocol, integrating the AWS Bedrock service, and utilizing PostgreSQL for vector storage.
This project is an AWS Bedrock Guardrails management server that provides functions for creating, editing, deleting, and exporting guardrails (including all policy types), and supports exporting configurations as Terraform files.
A knowledge base search MCP server project based on AWS Bedrock and Cloudflare Workers, supporting AI assistants to search for document content through the protocol.
This project demonstrates an AI - driven automation system based on AWS Bedrock, which can understand natural language support requests and automatically perform function management operations through an independent MCP server. It has enterprise - level security and complete testing coverage.