Fizzread

• Outcome: Review
• Latest audit: May 29, 2026 at 9:22 PM UTC
• Version:
  0.1.5

Overview

The skill does what it advertises, but it handles FizzRead API keys in a way that can expose a user's key in chat, command output, or logs.

Review before installing. Prefer the built-in demo key or set a revocable FizzRead key through a secure environment or secret mechanism. Do not let the agent print `FIZZREAD_API_KEY`, and avoid pasting a personal API key into chat unless you are comfortable with it remaining in the conversation context for that session.

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: 94% confidence
• Finding: The skill explicitly instructs the agent to accept a user-pasted API key, test it, and then remember and reuse it for subsequent requests in the session. Persisting sensitive credentials in conversational state without warning, minimization, or explicit consent increases the risk of unintended disclosure through logs, prompt leakage, debugging output, or cross-turn misuse.

VirusTotal

62/62 vendors flagged this skill as clean.