ClawHub

ADA-Predictor: Anti-Drug Antibody Risk Stratification

Security checks across malware telemetry and agentic risk

Overview

This is a local medical risk-scoring tool with no hidden access, network behavior, or persistence, but its treatment suggestions require clinician review.

Install only if you intend to use a local clinical risk calculator. Do not use its output as the sole basis for diagnosis, methotrexate dosing, biologic switching, or monitoring schedules; have a qualified clinician review any result and avoid entering identifiable patient data unless you control the environment.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
97% confidence
Finding
This skill generates patient-specific risk scores and treatment recommendations such as methotrexate optimization, therapeutic drug monitoring intervals, and alternative biologic selection, but it does not prominently warn that the output is only decision support and not a substitute for clinician judgment. In a medical context, users may over-trust the model's recommendations, creating a real risk of inappropriate treatment changes, delayed specialist evaluation, or unsafe care decisions.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The code generates individualized clinical risk scores and concrete treatment-monitoring recommendations such as methotrexate dosing guidance, therapeutic drug monitoring timing, and alternative mechanism-of-action suggestions, but presents them as authoritative output without any visible warning that this is a heuristic demonstration and not medical advice. In a healthcare context, omission of safety disclaimers and clinician-review requirements can cause users to rely on unvalidated outputs for treatment decisions, creating patient-safety risk even though there is no classic code-execution flaw.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal