Pharmaclaw Pharmacology Agent

• Outcome: Pass
• Latest audit: May 28, 2026 at 12:36 PM UTC
• Version:
  1.1.0

Overview

This skill is a local RDKit-based molecule profiling helper with no evidence of hidden access, persistence, credential use, network exfiltration, or destructive behavior.

Install only if you are comfortable running the bundled local Python script and have a trusted RDKit setup. Use the results for research triage only; do not rely on its ADME, toxicity-adjacent, interaction, BBB, or bioavailability estimates for medical, clinical, regulatory, or safety-critical decisions.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The agent returns pharmacology and ADME predictions as structured results with confidence values but no prominent disclaimer that they are heuristic, rule-based estimates not suitable for medical, clinical, or safety-critical decision-making. In a pharmacology context, this increases the risk that users or downstream agents over-trust the output and make hazardous decisions about bioavailability, BBB penetration, metabolism, or interaction risk.

VirusTotal

64/64 vendors flagged this skill as clean.