# pkotecha-eng/aria-mcp-server [Health: Active]

**Category:** Biology, Medicine and Bioinformatics  
**Repository:** https://github.com/pkotecha-eng/aria-mcp-server  
**GitHub Stars:** 3  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/pkotecha-eng-aria-mcp-server

## Description
Clinical research MCP server for trial coordinators and life sciences teams. Real-time access to PubMed (35M+ papers) and ClinicalTrials.gov (400K+ trials). No API key required.

## Tools
Capabilities this server exposes over MCP:

- **search_pubmed** — Search PubMed for peer-reviewed biomedical literature. Read-only operation. No authentication required. Uses NCBI E-utilities public API (rate limit: ~3 requests/sec). Returns up to 10 results per call. No pagination. Returns 'No papers found.' if no results match. Use for: research papers, drug mechanisms, clinical outcomes, disease studies, safety/efficacy data, biomarkers, diagnostics, and any scientific question.
- **search_clinical_trials** — Search ClinicalTrials.gov for clinical studies. Read-only operation. No authentication required. Uses ClinicalTrials.gov v2 public API (no rate limit documented). Returns up to 10 results per call. No pagination. Returns 'No clinical trials found.' if no results match. Use for: active trials, recruiting studies, inclusion/exclusion criteria, phase information, sponsor details, and trial locations.
- **search_isrctn** — Search the ISRCTN registry for UK and European clinical trials. Read-only operation. No authentication required. Complements ClinicalTrials.gov by covering trials at UK academic institutions and European research centers not listed on ClinicalTrials.gov. Returns up to 10 results per call, filtered for relevance. Returns 'No ISRCTN trials found.' if no results match. Use for: UK/European trials, academic institution studies, international coverage beyond ClinicalTrials.gov.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `uvx` (confidence: high):

```json
"mcpServers": {
  "aria-mcp-server": {
    "command": "uvx",
    "args": ["aria-mcp-server"]
  }
}
```

## Documentation

## What pkotecha-eng/aria-mcp-server MCP server does

The pkotecha-eng/aria-mcp-server MCP server connects an MCP-compatible Claude agent to three public biomedical data sources. It supports literature searches through PubMed, clinical-study searches through ClinicalTrials.gov, and additional UK and European trial searches through the ISRCTN registry. The server is intended for life sciences researchers, trial coordinators, drug discovery teams, and other users who need current evidence or study information in an agent workflow.

The server also provides three prompts: `clinical-research-brief` for combining evidence and trial information around a condition, `adverse-event-analysis` for organizing a drug’s safety profile, and `trial-eligibility-checker` for assessing eligibility across recruiting studies. Its resources include an ARIA capability and data-source overview, trial-phase references, high-impact journal lists by specialty, and FDA database references.

## How it works

The server runs locally as a Python MCP process and exposes its search functions to the connected client. `search_pubmed` uses the NCBI E-utilities public API to search biomedical literature. `search_clinical_trials` uses the version 2 public REST API from ClinicalTrials.gov. `search_isrctn` searches the ISRCTN registry for relevant UK and European studies that may not appear in ClinicalTrials.gov.

All three search tools are read-only and do not require authentication. Each returns at most 10 results per call, with no pagination support. PubMed requests are subject to an approximate rate limit of three requests per second. The material does not document a rate limit for ClinicalTrials.gov. Empty searches return source-specific messages such as `No papers found.` or `No clinical trials found.`

## Setup and configuration

Install the package with Python’s package installer, then start the server using its installed command:

```bash
pip install aria-mcp-server
aria-mcp-server
```

The documented Claude Desktop configuration uses the `aria-mcp-server` executable as a local MCP server command:

```json
{
  "mcpServers": {
    "aria-clinical-research": {
      "command": "aria-mcp-server"
    }
  }
}
```

Python 3.12 or newer is listed in the project’s technology stack. No API key or other environment variable is required according to the provided material.

## Tools and capabilities

- Search up to 35 million PubMed papers for biomedical literature, including disease studies, mechanisms, outcomes, biomarkers, diagnostics, and safety or efficacy evidence.
- Search more than 400,000 ClinicalTrials.gov studies for recruiting status, phases, sponsors, locations, and eligibility information.
- Search ISRCTN for UK and European trials, including studies associated with academic institutions and research centers.
- Use structured prompts for research briefs, adverse-event analysis, and trial eligibility checks.
- Read reference resources covering trial phases, selected journals, FDA databases, and ARIA data sources.

## Limitations and notes

The pkotecha-eng/aria-mcp-server MCP server only performs read-only searches; the provided tools do not create, update, or manage records in the source registries. Results are capped at 10 per tool call, and none of the listed search tools supports pagination, so broad research may require separate refined queries.

PubMed access uses NCBI E-utilities and should be used within its approximate three-request-per-second limit. ClinicalTrials.gov’s rate-limit policy is not documented in the supplied material. ISRCTN is described as complementary coverage for UK and European trials rather than a replacement for the other registries. The server provides research information and search results, but the material does not establish that its output constitutes medical advice or a substitute for professional clinical judgment.

_Full upstream README: https://allmcps.com/mcp/pkotecha-eng-aria-mcp-server/readme_