# AWS HealthOmics

Connects AI assistants to AWS HealthOmics to launch genomic workflows, validate WDL or CWL code, track tasks, and query CloudWatch OpenTelemetry metrics.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 9939 repo stars
- Author: awslabs
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/awslabs/mcp/tree/HEAD/src/aws-healthomics-mcp-server
- Updated: 2026-10-23
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/aws-healthomics

## What it does

This server connects coding assistants to AWS HealthOmics for managing genomic workflows. It allows models to create, lint, and bundle pipelines written in WDL, CWL, or Nextflow. Workflow validation relies on `miniwdl` and `cwltool`. Tools can also locate genomics data across S3 buckets and sequence stores, pairing BAM files with BAI indexes automatically.

Beyond authoring, it handles execution tracking and failure diagnosis. An agent can launch runs with custom VPC networking, inspect engine logs, and evaluate hardware bottlenecks across CPU, memory, and I/O. Fine-grained telemetry is retrieved via the CloudWatch PromQL API. Because metrics sample roughly every 30 seconds, tasks that finish in under 60 seconds produce no datapoints, leaving run manifests as the only fallback for quick steps.

## Requirements
- Needs: Docker
- Works with: AWS

## Install

### Claude Code (native platform)

**Install**
- The MCP registry has no launch command for it: see the repository README

## When to use it
- Lint WDL and CWL workflow files using `miniwdl` and `cwltool`
- Launch genomics workflow runs in AWS HealthOmics with custom VPC settings
- Discover and group paired FASTQ files or BAM/BAI indexes in S3
- Diagnose failed workflow runs through engine and task logs
- Compare CPU and memory resource consumption across separate runs

## Example prompts
- Lint this WDL workflow file before deploying it to HealthOmics.
- Search the S3 bucket for paired FASTQ files and associated indexes.
- Inspect the task logs of my latest failed HealthOmics run and pinpoint the error.

## Similar
- [dnanexus-integration](https://affpapa.org/en/nevibecode/skills/dnanexus-integration.md) — skill · 2100 installs
- [bioinformatics-workflows](https://affpapa.org/en/nevibecode/skills/bioinformatics-workflows.md) — skill · 155 installs
- [AWS Docs and Regions](https://affpapa.org/en/nevibecode/mcp/aws-docs-and-regions.md) — MCP server · 6621 installs
- [AWS Marketplace](https://affpapa.org/en/nevibecode/mcp/aws-marketplace.md) — MCP server · 655 installs
- [aws-serverless](https://affpapa.org/en/nevibecode/skills/aws-aws-serverless.md) — skill · 11600 installs
- [aws-iam](https://affpapa.org/en/nevibecode/skills/aws-iam.md) — skill · 11500 installs

From the same set: [awslabs/mcp](https://affpapa.org/en/nevibecode/by/awslabs/mcp.md) — 36 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
