发芽报告

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:34 PM UTC
• Version:
  1.0.0

Overview

This is an instruction-only analysis skill that asks for recordings, transcripts, or topics to generate a structured report, with no code, installs, credentials, persistence, or hidden data movement found.

Before using this with private recordings or conversations, confirm you have consent from participants and understand how your agent or host transcribes and stores audio. Prefer providing a transcript or redacted text when the content is sensitive.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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

Findings (3)

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill explicitly invites users to upload audio recordings for transcription, which commonly contain sensitive personal data, but it provides no notice about privacy, consent, retention, or third-party processing. This creates a real privacy and compliance risk because users may submit recordings involving themselves or others without understanding how that data will be handled.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The skill description is broad enough to overlap with many ordinary requests about analysis, recordings, conversations, or brainstorming. That can cause unintended invocation, leading the agent to apply this skill when the user did not explicitly want deep analysis or transcription, which may expose sensitive content to unnecessary processing and produce off-target behavior.

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The '何时使用' section lists very general situations but does not define boundaries, exclusions, or examples of when the skill should not activate. In practice, this increases accidental activation on normal discussion, ideation, or summary requests, which can cause over-collection of user content and unexpected transformation of input into a detailed report.

VirusTotal

64/64 vendors flagged this skill as clean.