NoahAI clinical-trial query

Security checks across malware telemetry and agentic risk

Overview

This is a coherent clinical-trial search skill that uses a Noah API token to send trial-search terms to Noah's service, with privacy cautions but no hidden or destructive behavior found.

Install this only if you trust Noah's API service with your clinical-trial search terms and have a revocable NOAH_API_TOKEN. Avoid including unnecessary personal medical details, confirm before running it on ambiguous health questions, and use --params-file or --output only with files you intentionally chose.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 92% confidence
Finding: The skill declares runtime requirements and documents execution of a Python script that reads environment variables, performs network access, and writes query files, but it does not declare corresponding permissions in a first-class permission model. This creates a transparency and governance gap: the agent or reviewer may underestimate what the skill can access, especially the API token in NOAH_API_TOKEN and any locally written parameter files.

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The trigger language is very broad and covers many common biomedical concepts, which can cause the skill to activate for loosely related health questions rather than only explicit clinical-trial lookup requests. Over-broad routing can expose user queries and extracted medical terms to an external API unnecessarily, increasing privacy leakage and the chance of unintended external actions.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal