script-breakdown

Pass

Audited by Gen Agent Trust Hub on Nov 21, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data from YouTube video content which could potentially contain malicious instructions.
  • Ingestion points: The skill retrieves video transcripts, titles, and channel descriptions using the TubeLab MCP tool as part of the analysis workflow (SKILL.md).
  • Boundary markers: There are no explicit instructions to use delimiters or ignore instructions that may be embedded within the external transcripts or descriptions.
  • Capability inventory: The skill does not possess high-risk capabilities such as arbitrary file system access, network exfiltration, or code execution; it focuses on text analysis and output generation.
  • Sanitization: The instructions do not specify any sanitization, filtering, or validation steps for the content fetched from YouTube before it is processed.
  • Remediation: Consider wrapping external transcript content in clear boundary markers (e.g., XML tags or triple backticks) and adding instructions to the agent to disregard any commands or directives found within the ingested data.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 21, 2026, 04:16 PM