wegene-oauthLABSCO SUMMARY
Your own genetic reports, readable in the place where you can ask questions about them.
The flow is short: authorise once, pick the profile, check what reports exist, then pull the one you want. Results arrive as the raw JSON from WeGene's API, so the model is reading vendor fields rather than a prepared summary — and the browser authorisation has to be finished within 120 seconds or the step starts again.
What it is
A server over the WeGene Open API: authorise in the browser, list the profiles on your account, see which reports exist, then fetch the results of one of them.
What you get
- A browser OAuth flow started from the conversation, completed within 120 seconds
- The profiles on your WeGene account, by name and id
- The report catalogue — names, descriptions and endpoints — so the right report can be picked
- The result JSON for one report on one profile, as WeGene's API returns it
- Every authorised report also exposed as a wegene:// resource with its own name and description
Requirements
A WeGene account with reports on it, plus a WeGene Open API key and secret — copy .env.example to .env and fill them in. Runs with uv from a clone.
Setup effort
Build from source — git clone https://github.com/xraywu/mcp-wegene-assistant.git
USE CASES
Reading a genetic report without decoding the JSON yourself
Find the report's endpoint in the catalogue, then fetch its results and have them explained in plain language.
Authorising access the first time
Start the browser OAuth flow from the conversation and complete it inside the 120-second window.
Choosing whose data to look at on a family account
List the profiles by name and id, then fetch reports against the one you meant.
Finding out which reports your account actually has
Read the report meta info — names, descriptions and endpoints — before requesting any results.
Comparing the same report across two profiles
Take the profile ids, then fetch the same report endpoint for each and read them side by side.
4 tools
Listed in project docs. Your client may show a different list.
wegene-get-profileswegene-get-report-infowegene-get-reportBEFORE YOU INSTALL
What to know first
Tools you get
4 tools named in the project's documentation.
Requires
- uv on PATH
- Python 3.12 or later
- A WeGene Open API key and secret, in the project's .env file
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
git clone https://github.com/xraywu/mcp-wegene-assistant.git
cd mcp-wegene-assistant
uv sync --dev --all-extras
cp .env.example .envEdit .env — put your WeGene Open API key and secret in it
MCP Clients
Claude Desktop
{
"mcpServers": {
"wegene-assistant": {
"command": "uv",
"args": [
"--directory",
"<YOUR_PATH_TO_WEGENE_ASSISTANT>",
"run",
"wegene-assistant"
]
}
}
}<YOUR_PATH_TO_WEGENE_ASSISTANT> is the full path of your mcp-wegene-assistant clone — the folder holding the .env file from the steps above.