OakVar

OakVar MCP Server lets you control OakVar’s genomic variant analysis capabilities through AI assistants. By exposing OakVar’s modules and workflows via the MCP protocol, you can run annotations, generate reports, manage modules, and query results using natural language requests from MCP clients like Claude Desktop or ChatGPT Desktop.

How to use

You connect an MCP client to the OakVar MCP Server and issue high-level requests. Use questions or commands such as asking which OakVar modules are installed, installing a specific annotator, running an analysis on your data, or filtering and exporting results. The server translates your natural-language requests into OakVar actions, executes the analysis, and returns the results through the MCP client in a convenient format.

How to install

Prerequisites: ensure Python is installed on your system. You may also use a virtual environment to isolate dependencies.

Step 1: Install OakVar and the MCP server.

pip install oakvar oakvar-mcp

Step 2: Set up OakVar for first use.

ov system setup

Step 3: Configure your MCP client to connect to the OakVar MCP Server. You will typically add a server entry in your MCP client’s config that points to the OakVar MCP executable.

Configuration and usage notes

The MCP server is intended to be run in the same Python environment as OakVar. If you use a virtual environment, point your MCP client to the full path of that environment’s OakVar MCP executable when configuring the client.

# Claude Desktop configuration block example
{
  "mcpServers": {
    "oakvar": {
      "command": "oakvar-mcp"
    }
  }
}

# ChatGPT Desktop configuration block example
{
  "mcpServers": {
    "oakvar": {
      "command": "oakvar-mcp"
    }
  }
}

Notes and tips

There is a development flow for building the MCP server locally: clone the repository, install in editable mode, and run tests when you want to validate changes.

Available tools

oakvar_version

Get OakVar version

oakvar_system_check

Verify installation integrity

oakvar_system_setup

Configure OakVar for initial use

oakvar_modules_dir

Get or set the modules directory

oakvar_module_list

List available modules

oakvar_module_info

Get details for a specific module

oakvar_module_install

Install modules into OakVar

oakvar_module_uninstall

Remove installed modules

oakvar_module_update

Update modules to latest versions

oakvar_run

Run the annotation pipeline on provided data

oakvar_report

Generate reports from results in various formats

oakvar_sqliteinfo

Get information about the OakVar SQLite databases

oakvar_filtersqlite

Filter databases based on criteria

oakvar_query

Execute SQL queries against OakVar data sources

oakvar_new_module

Create templates for new modules

oakvar_new_exampleinput

Create test input data for modules

oakvar_module_pack

Pack modules for distribution

oakvar_store_fetch

Refresh the module store cache

oakvar_store_register

Register modules with the store