exon-research/genomi

Local-first, open-source Claude Science alternative, before Claude Science is a thing. Turn your AI agent into personal DNA expert.

What it solves

Genomi provides a way for AI agents to analyze personal genomic data privately and accurately. It solves the problem of genomic data being too vast and complex for humans or general-purpose AI to process without hallucinating or losing detail, by providing a grounded, local-first workspace for DNA analysis.

How it works

Genomi acts as an AI agent runtime using the Model Context Protocol (MCP). It parses raw DNA files (such as VCF, gVCF, or exports from providers like 23andMe) into a local "Active Genome Index" (AGI). The AI agent then uses this index and a set of public genetics evidence libraries to query specific variants, check pharmacogenomics, and research disease risks without the raw DNA ever leaving the user's machine.

Who it’s for

It is designed for individuals who have raw genomic data and want to explore their genetic predispositions, medication responses, and inherited traits using an AI agent, as well as researchers using the developer preview of GenomiLab for disease investigation.

Highlights

  • Local-first privacy: Raw genome files remain on the user's machine.
  • MCP-native: Compatible with any MCP-capable host (e.g., Claude Code, Cursor, Windsurf).
  • Broad source support: Supports a wide range of formats including VCF, gVCF, FASTQ, and direct-to-consumer exports from 23andMe, AncestryDNA, and others.
  • Evidence-backed answers: Uses a local evidence library to ensure agents provide cited answers rather than guessing from memory.
  • GenomiLab: A developer preview application for structured disease investigation and patient-molecular profiling.

Related

  • Dispatch
    Inside Genebench-Pro: Case Studies in Genomic AI Benchmarking

    OpenAI's Genebench-Pro provides a specialized benchmark for evaluating AI models on complex genomic tasks, featuring ten detailed case studies across somatic oncology, functional genomics, and population genetics.

  • Project
    snap-stanford/Biomni

    Biomni is a general-purpose biomedical AI agent that automates research tasks and hypothesis generation by combining LLM reasoning, retrieval-augmented planning, and code execution.

  • Project
    Open-Curiosity/gini-agent

    A personal AI agent that uses interactive inline controls and live browser handoffs to solve the problem of human-in-the-loop requirements for complex tasks.

  • Project
    modelscope/ms-agent

    MS‑Agent is an open‑source Python framework that lets you build, run, and manage autonomous AI agents. It provides a customizable harness for planning, tool execution, permission checks, and feedback, with built‑in context‑compression, long‑term memory, and DAG‑style multi‑agent workflows. Agents can be used via a Web UI, terminal UI, CLI, or embedded SDK, and the ecosystem includes reusable “skills”, multimodal input, and ready‑made applications for research, code generation, video creation, and finance.

  • Project
    Syngnat/GoNavi

    GoNavi is a lightweight, cross‑platform desktop database client built with Go + Wails and React. It supports dozens of data sources, provides a schema‑aware AI assistant (MCP) that can generate and explain SQL via OpenAI‑compatible models, and offers a virtualised grid for large result sets. Installers are ~20‑26 MB, and a separate CLI enables headless scripting.