SkillAgentSearch skills...

Genomic Agent Discovery

AI agents that collaborate to analyze your DNA. 12 public databases, real-time dashboard, runs 100% locally. Drop in your 23andMe, AncestryDNA, or VCF file.

Install / Use

/learn @HelixGenomics/Genomic Agent Discovery

README

<p align="center"> <img src="docs/helix-logo.png" alt="Helix Genomics Agents" width="120"> </p> <h1 align="center">Genomic Agent Discovery</h1> <p align="center"> <strong>AI agents that collaborate to analyze your DNA. Open source. Runs locally. Your data never leaves your machine.</strong> </p> <p align="center"> <a href="#quick-start">Quick Start</a> &bull; <a href="#dashboard">Dashboard</a> &bull; <a href="#presets">Presets</a> &bull; <a href="#agent-prompts">Agent Prompts</a> &bull; <a href="#architecture">Architecture</a> &bull; <a href="#configuration-reference">Configuration</a> &bull; <a href="#database">Database</a> &bull; <a href="#privacy--security">Privacy</a> &bull; <a href="#about-this-project">About This Project</a> </p> <p align="center"> <img alt="License: MIT" src="https://img.shields.io/badge/license-MIT-blue.svg"> <img alt="Node 18+" src="https://img.shields.io/badge/node-%3E%3D18-brightgreen.svg"> <img alt="MCP Compatible" src="https://img.shields.io/badge/MCP-compatible-blueviolet.svg"> <img alt="Databases" src="https://img.shields.io/badge/databases-12%2B-orange.svg"> <img alt="PRs Welcome" src="https://img.shields.io/badge/PRs-welcome-brightgreen.svg"> </p>

Upload your raw DNA file from 23andMe, AncestryDNA, MyHeritage, FamilyTreeDNA, or any VCF -- and watch a team of AI agents fan out across 12+ public genomics databases, share discoveries with each other in real time, and produce a comprehensive health report. Everything runs on your machine. Nothing is uploaded anywhere.

<p align="center"> <img src="docs/screenshots/gif-pipeline-running.gif" alt="Pipeline running — agents analyzing, findings arriving, inter-agent chat" width="720"> <br> <em>Real-time pipeline: 7 agents collaborating on a cancer genomics analysis</em> </p>

Quick Start

Have a Claude Pro or Max subscription? (Recommended)

No API key needed. Your subscription covers everything.

# 1. Install Claude CLI if you haven't already
npm install -g @anthropic-ai/claude-code

# 2. Log in once — opens browser OAuth (free, uses your Claude Pro/Max subscription)
claude login

# 3. Clone, build, and run
git clone https://github.com/HelixGenomics/Genomic-Agent-Discovery.git
cd Genomic-Agent-Discovery
npm install && npm run build-db
npm start -- --dna ~/Downloads/my-dna-raw.txt

A dashboard opens in your browser and you can watch the agents work. That's it — no API keys, no per-token charges.

Dashboard-first mode: Want to configure everything in the browser first?

npm start -- --serve
# Opens http://localhost:3000 — select your DNA file, pick a preset, customize agents, then click Start

Using an Anthropic API key instead

export ANTHROPIC_API_KEY=sk-ant-...   # get one at console.anthropic.com

git clone https://github.com/HelixGenomics/Genomic-Agent-Discovery.git
cd Genomic-Agent-Discovery
npm install && npm run build-db
npm start -- --dna ~/Downloads/my-dna-raw.txt --provider anthropic-api

Typical cost: $1–5 per analysis run depending on preset. See Provider options for OpenAI, Gemini, Ollama, and others.

Dashboard

The dashboard is a real-time mission control for your genomic analysis. It provides full visibility into agent status, findings, inter-agent communication, and costs — all in a single page.

Setup Panel

When you launch the dashboard, you'll see the setup panel where you configure your analysis before starting.

<p align="center"> <img src="docs/screenshots/01-setup-panel.png" alt="Dashboard setup panel" width="720"> <br> <em>Setup panel — select a preset, configure settings, and start your analysis</em> </p>

Preset Selection

Choose from 6 built-in research presets, each tuned for a specific domain. Selecting a preset instantly configures the agent pipeline, prompts, models, and focus areas.

<p align="center"> <img src="docs/screenshots/-preset-switching.gif" alt="Switching between presets" width="720"> <br> <em>Switch between presets — each configures a different agent team with specialized prompts</em> </p>

Available presets:

| Preset | Agents | Est. Cost | Focus | |--------|--------|-----------|-------| | Quick Scan ⚡ | 2 | $0.05–0.10 | Fast overview across all domains | | Cancer Research 🔬 | 7 | $0.50–2.00 | Deep cancer & tumor genetics with DPYD safety, platinum chemo, immunotherapy, and targeted therapy agents | | Cardiovascular ❤️ | 6 | $0.30–1.00 | Lipid genetics, arrhythmia risk, coagulation, and structural heart | | Pharmacogenomics 💊 | 4 | $0.20–0.80 | CYP enzyme panel, drug transporters, and full CPIC pharmacogene coverage | | Rare Disease 🧬 | 7 | $0.40–1.50 | Metabolic disorders, neurological conditions, connective tissue, immunodeficiency, and rare cancer syndromes | | Custom ⚙️ | You decide | Varies | Build your own agent pipeline from scratch |

Database Status

The setup panel shows a live view of your annotation databases — which are loaded, how many rows each contains, and the total database size. This tells you at a glance whether you need to run npm run build-db.

<p align="center"> <img src="docs/screenshots/03-database-status-expanded.png" alt="Database status panel expanded" width="720"> <br> <em>13 databases loaded — 8.4M total rows across ClinVar, GWAS, CPIC, AlphaMissense, and more</em> </p>

Editable Agent Prompts & Tier Grouping

Every preset shows its agents grouped by tier: Collection (cheap models, high tool calls), Synthesis (smarter models combining findings), and Report (final output). Click any agent to expand and edit its prompt, change its model, or adjust settings.

<p align="center"> <img src="docs/screenshots/05-editable-agent-prompts.png" alt="Editable agent prompts with tier grouping" width="720"> <br> <em>Cancer preset — 5 haiku collectors, 1 sonnet synthesizer, 1 haiku report writer. All prompts editable.</em> </p> <p align="center"> <img src="docs/screenshots/06-tier-grouping.png" alt="Tier grouping showing collection, synthesis, and report phases" width="720"> <br> <em>Tiered pipeline: cheap models do high-volume database queries, expensive models synthesize findings</em> </p>

Template Import & Export

Share your custom agent configurations as JSON template files. Export your current setup (including any prompt edits) and import templates shared by others.

<p align="center"> <img src="docs/screenshots/04-template-import-export.png" alt="Template import and export buttons" width="720"> <br> <em>Import/Export buttons below the preset selector — share templates as JSON files</em> </p>

Templates include all agents, prompts, model assignments, and settings. An example Debendox/Trisomy 9 investigation template is included in config/templates/.

Output Configuration

Toggle markdown output and set a shared output directory for all agent reports. Files are named automatically based on agent IDs (e.g., cancer-collector.md, synthesizer.md).

<p align="center"> <img src="docs/screenshots/07-output-config.png" alt="Output directory configuration" width="720"> <br> <em>Output config — single shared directory, file preview chips show what will be generated</em> </p>

The default output directory is MD_DOCS/ in your repo root. Edit the path to save anywhere. Each agent writes its findings to a separate markdown file.

Pipeline Animation

Once the analysis starts, the dashboard shows a real-time canvas visualization of the agent pipeline. Agents are distributed across concentric rings (scales to 20+ agents), with animated connections showing data flow and collaboration.

<p align="center"> <img src="docs/screenshots/gif-pipeline-running.gif" alt="Full pipeline animation" width="720"> <br> <em>Live pipeline — agents spawn, run, share findings, chat with each other, and complete</em> </p>

The pipeline view shows:

  • Agent status — spawning (blue pulse), running (green glow), done (solid green), error (red)
  • Findings — each discovery appears in real-time with gene, confidence, and clinical category
  • Inter-agent chat — agents coordinate in real-time (e.g., cancer agent alerts pharma agent about DPYD variant)
  • Cost tracking — estimated cost updates as agents consume tokens
  • Log sizes — see how much each agent has written

Full Configuration Walkthrough

<p align="center"> <img src="docs/screenshots/gif-config-walkthrough.gif" alt="Full configuration walkthrough" width="720"> <br> <em>Complete walkthrough of the setup panel — presets, prompts, output, settings, and launch</em> </p>

What You Get

A structured genomic health report covering cancer genetics, cardiovascular risk, pharmacogenomics (how you metabolize 100+ drugs), neurological traits, and metabolic health -- all cross-referenced across 12 public databases and prioritized by clinical significance.

A real-time dashboard where you can watch agents query your DNA, discover findings, send messages to each other, and build on each other's research. It looks like a mission control room for your genome.

Raw findings in JSON for downstream analysis, integration with other tools, or building your own visualizations.

What the agents actually do

  1. Parse your raw DNA file (600K-5M+ variants depending on source)
  2. Query each variant against ClinVar, GWAS Catalog, AlphaMissense, CADD, PharmGKB, CIViC, and more
  3. Talk to each other -- the cancer agent might tell the pharma agent "this patient has a DPYD variant, check fluorouracil metabolism"
  4. Deduplicate automatically so you don't get the same finding five times
  5. Synthesize cross-domain patterns a single agent would miss
  6. Write a clear, readable report with appropriate medical disclaimers

Supported DNA Files

| Format | Provider | Typical Variants | File Extension | |--------|----------|-----------------|----------------| | 23andMe | 23andMe | ~6

Related Skills

View on GitHub
GitHub Stars7
CategoryDevelopment
Updated2h ago
Forks1

Languages

JavaScript

Security Score

90/100

Audited on Apr 1, 2026

No findings