A bioinformatics-native AI agent skill library where each skill is a runnable, reproducible analysis for genomics tasks, built on OpenClaw.
aws-samples/amazon-bedrock-agents-healthcare-lifesciences
A toolkit of healthcare and life sciences reference agents, reusable domain skills, and MCP tools for AI coding assistants and AWS deployments.
- Stars
- 275
- Forks
- 115
- Open issues
- 97
- Last commit
- 9 days ago
What is it?
This AWS sample repository collects healthcare and life sciences (HCLS) AI capabilities in three forms: a catalog of more than 36 reference agents covering drug research, clinical trials, genomics and commercialization, reusable skills that encode domain workflows, and MCP servers that expose biomedical databases and ontologies as callable tools. The agents are built with the Strands framework and designed for deployment on Amazon Bedrock AgentCore.
The main change in direction is making the same domain knowledge usable without deploying a full agent. Skills can be installed into AI coding assistants such as Claude Code and Kiro, and MCP servers can be connected to end-user platforms like Amazon Quick, so researchers can query tools like UniProt, HPO or PubMed in natural language. MCP servers are grouped by deployment model: AgentCore Gateway, AgentCore Runtime, pre-existing AWS servers and third-party public servers.
Some agents depend on shared infrastructure such as Redshift and a Bedrock Knowledge Base, which the repository provides as a CloudFormation stack. The project is explicitly for demonstration only and is not intended for clinical use or production handling of protected health information without a BAA.
Key features
- 01Catalog of 36+ reference agents built with the Strands framework for AgentCore deployment
- 02Reusable skills covering genomics, drug discovery, clinical trials, biomarkers, research and terminology
- 03MCP servers exposing biomedical databases and ontologies such as Biomni, OLS, PubMed and Open Targets
- 04Setup guides for Claude Code, Kiro and Amazon Quick
- 05CloudFormation stack for shared infrastructure including Redshift and a Bedrock Knowledge Base
Use cases
- Letting developers in Claude Code or Kiro query UniProt, HPO or disease-gene associations in natural language
- Connecting biomedical literature and ontology tools to Amazon Quick for researchers
- Using the agent templates as a starting point for building HCLS agents on Bedrock AgentCore
- Studying reference implementations for genomics or clinical trial workflows
Quick start
- 1
Clone the repository
git clone https://github.com/aws-samples/amazon-bedrock-agents-healthcare-lifesciences.git - 2
Enter the repository folder
cd amazon-bedrock-agents-healthcare-lifesciences - 3
Open the platform-specific setup guide for Claude Code or Kiro
cat platforms/claude-code/README.md - 4
Follow the guide to install skills and connect MCP servers
Things to know
- This solution is for demonstration only and is not for clinical use or production.
- Some agents need AWS infrastructure such as Redshift and a Bedrock Knowledge Base, which requires a CloudFormation deployment.
- The repository is being repositioned, so its structure and catalog may change.
In collections
Topics
FAQ
- What is amazon-bedrock-agents-healthcare-lifesciences?
- A toolkit of healthcare and life sciences reference agents, reusable domain skills, and MCP tools for AI coding assistants and AWS deployments. This AWS sample repository collects healthcare and life sciences (HCLS) AI capabilities in three forms: a catalog of more than 36 reference agents covering drug research, clinical trials, genomics and commercialization, reusable skills that encode domain workflows, and MCP servers that expose biomedical databases and ontologies as callable tools. The agents are built with the Strands framework and designed for deployment on Amazon Bedrock AgentCore.
- Is amazon-bedrock-agents-healthcare-lifesciences free to use? What license does it have?
- amazon-bedrock-agents-healthcare-lifesciences is open source under the MIT-0 license. Check the license terms before using it in commercial products.
- Is amazon-bedrock-agents-healthcare-lifesciences still maintained?
- Actively maintained. The last commit to amazon-bedrock-agents-healthcare-lifesciences was on Oct 29, 2026.
- What can I use amazon-bedrock-agents-healthcare-lifesciences for?
- Letting developers in Claude Code or Kiro query UniProt, HPO or disease-gene associations in natural language Connecting biomedical literature and ontology tools to Amazon Quick for researchers Using the agent templates as a starting point for building HCLS agents on Bedrock AgentCore Studying reference implementations for genomics or clinical trial workflows
- How do I install amazon-bedrock-agents-healthcare-lifesciences?
- Clone the repository: git clone https://github.com/aws-samples/amazon-bedrock-agents-healthcare-lifesciences.git Enter the repository folder: cd amazon-bedrock-agents-healthcare-lifesciences Open the platform-specific setup guide for Claude Code or Kiro: cat platforms/claude-code/README.md Follow the guide to install skills and connect MCP servers
- Where can I get amazon-bedrock-agents-healthcare-lifesciences?
- The source code and installation instructions are on GitHub: https://github.com/aws-samples/amazon-bedrock-agents-healthcare-lifesciences
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A deep learning variant caller that identifies genetic variants from aligned next-generation DNA sequencing reads and outputs VCF/gVCF files.
Python tool that predicts a person's genetic ancestry from direct-to-consumer SNP data by comparing it against 1000 Genomes samples.
A self-hosted AI health data engine that standardizes lab reports, wearable data and genetic files into one traceable record, with an MCP server.
A collection of open source MCP servers that give AI assistants access to AWS documentation, services, and tooling.
Official AWS-supported MCP servers, skills, and plugins that help AI coding agents build, deploy, and manage applications on AWS.