Percy

• Outcome: Review
• Latest audit: May 29, 2026 at 9:37 PM UTC
• Version:
  1.0.2

Overview

This Percy skill is a coherent integration, but it gives an agent broad authenticated Percy API access, including write and delete requests, without explicit confirmation guidance.

Install only if you are comfortable letting an agent operate through your authenticated Percy connection. Prefer prebuilt Membrane actions, use the least-privileged Percy account available, and require explicit approval before any POST, PUT, PATCH, or DELETE proxy request.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• 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: 88% confidence
• Finding: The skill instructs the agent to use direct proxy requests with full HTTP method support, including POST, PUT, PATCH, and DELETE, without emphasizing that these operations can change or delete live Percy data. In an agent setting, this increases the chance of unintended state-changing actions against a remote service, especially if the agent falls back to raw requests without explicit user confirmation.

VirusTotal

65/65 vendors flagged this skill as clean.