AACT Clinical Trials MCP
Clinical Trial Data SQL Query MCP Server
Install & Use
Copy this prompt and send it to your AI assistant (Claude / Cursor / TRAE / Codex / WorkBuddy etc.) to auto-install:
Help me install this AI Skill: AACT Clinical Trials MCP. It is used for: Clinical Trial Data SQL Query MCP Server Full Skill content: https://321skill.com/skills/aact-clinical-trials-mcp/raw/index.md Read that page and install it.
The prompt includes a link to the full Skill content. You can also view the full content.
This Skill addresses the inefficiency in clinical trial data analysis. Traditionally, fetching data via the ClinicalTrials.gov API returns JSON records one by one, making it difficult to perform cross-trial aggregated statistics, such as calculating the average duration for a certain type of trial or grouping counts by sponsor. In contrast, the AACT database organizes the data into over 70 structured PostgreSQL tables, enabling direct SQL queries for complex analysis and significantly improving efficiency.
Usage is straightforward: you simply register for a free account on the AACT website to obtain database credentials, then install this MCP server via the Claude Desktop plugin, uvx, Docker, or similar methods. After configuring the username and password, you can use natural language to instruct the AI to execute SQL queries—for example, "List the top 10 sponsors and their status distribution for Phase 2 NSCLC trials from 2020 to 2025." The server provides tools such as database_info, list_tables, describe_table, get_column_values, search_columns, read_query, and fetch_rows, with support for paginated result browsing.
It is well-suited for teams or individuals requiring in-depth analysis of clinical trial data, particularly drug development researchers, medical statisticians, academic investigators, and pharmaceutical market analysts monitoring competitive pipelines. For users already familiar with SQL but who prefer not to write complex queries manually, the AI can automatically generate and execute SQL, lowering the barrier to entry.
Before analysis, it is recommended to first use list_tables and describe_table to understand the table structure, then use search_columns to locate specific fields. Note that the AACT database is not updated in real-time; it is typically synchronized with ClinicalTrials.gov data on a weekly basis, making it unsuitable for scenarios requiring the latest registration information. Additionally, all queries are read-only and will not modify the database.
Key Features
Unlike the official ClinicalTrials.gov API, the AACT MCP server connects directly to the PostgreSQL database, supporting complex analyses—such as aggregation, grouping, and sorting across trials—with a single SQL statement, eliminating the need to parse JSON records individually and perform calculations on the client side.
Limitations
Requires registering for a free AACT account and configuring database credentials; the database is synchronized weekly and is not updated in real-time; queries rely on PostgreSQL syntax and do not support non-SQL operations.
FAQ
Is an API key required?
No API key is required, but you need to register for an account on the AACT website to obtain a database username and password.
What is the data update frequency?
The AACT database synchronizes with ClinicalTrials.gov data on a weekly basis, typically updated every Monday.
Installation guide for AI assistants
If your AI coding assistant (Claude Code, Cursor, TRAE etc.) can see this page, send it this message to auto-install:
Visit https://321skill.com/skills/aact-clinical-trials-mcp/raw/index.md to read the original Skill definition (Markdown format) for AACT Clinical Trials MCP, and install it according to the instructions.
Raw Markdown URL for AI: /skills/aact-clinical-trials-mcp/raw/index.md