- Home
- MCP servers
- AACT
AACT
- python
16
GitHub Stars
python
Language
6 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"navisbio-aact_mcp": {
"command": "uvx",
"args": [
"mcp-server-aact"
],
"env": {
"DB_USER": "USERNAME",
"DB_PASSWORD": "PASSWORD"
}
}
}
}This MCP server provides direct access to the AACT clinical trials data via the MCP framework. You can query database tables, inspect their structures, and run read queries to retrieve clinical trial information for AI assistants and analytical workflows.
How to use
You interact with the AACT MCP server through an MCP client by performing three core actions: list tables to understand the database structure, describe a specific table to see its columns and data types, and read queries to execute safe SELECT statements against the data. Use these tools to fetch trial identifiers, titles, enrollment numbers, and other attributes across ClinicalTrials.gov data.
Practical usage patterns include exploring what data is available, understanding the table schemas before querying, and running read queries with constraints to retrieve exactly the fields you need. For example, you can list the available tables first, then describe the most relevant table (such as studies) to confirm column names, and finally execute a SELECT-like read to pull specific fields for analysis.
How to install
Prerequisites you need to have before starting: you will configure one of the MCP server options shown below, and you must have access to a database user for AACT.
Step 1. Create a free account to access the AACT database credentials.
Step 2. Obtain the database credentials and set environment variables for your MCP runtime.
Step 3. Use one of the provided configurations to start the MCP server.
Option A: Run with the published package using the UVX runtime (no build required) and provide your DB credentials through environment variables.
Option A: Using the published package with UVX
{
"mcpServers": {
"CTGOV-MCP": {
"command": "uvx",
"args": [
"mcp-server-aact"
],
"env": {
"DB_USER": "USERNAME",
"DB_PASSWORD": "PASSWORD"
}
}
}
}
Option B: Using Docker
{
"mcpServers": {
"CTGOV-MCP-DOCKER": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"--env", "DB_USER=YOUR_USERNAME",
"--env", "DB_PASSWORD=YOUR_PASSWORD",
"navisbio/mcp-server-aact:latest"
]
}
}
}
Option C: Running from source (development) with Docker
{
"mcpServers": {
"CTGOV-MCP-DOCKER": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"--env", "DB_USER=YOUR_USERNAME",
"--env", "DB_PASSWORD=YOUR_PASSWORD",
"navisbio/mcp-server-aact:latest"
]
}
}
}
Troubleshooting
If you see the error spawn uvx ENOENT, it means the uvx command cannot be found at its expected location. This can happen if uvx is installed in a non-standard path. Update your configuration to point to the full path of the uvx executable.
{
"mcpServers": {
"CTGOV-MCP": {
"command": "/Users/username/.local/bin/uvx",
"args": [
"mcp-server-aact"
],
"env": {
"DB_USER": "USERNAME",
"DB_PASSWORD": "PASSWORD"
}
}
}
}
Available tools
list_tables
Get an overview of all available tables in the AACT database to understand its structure.
describe_table
Examine the detailed structure of a specific AACT table, including column names and data types.
read_query
Execute a SELECT query on the AACT database with validation to safely retrieve data using SQL.