hook-writing
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides structured guidance for creating YouTube content and does not demonstrate any patterns associated with prompt injection, data exfiltration, or remote code execution.
- [INDIRECT_PROMPT_INJECTION]: The skill references an external project file (.agents/youtube-context.md) to gather context for the video's tone and audience. Ingestion points: .agents/youtube-context.md (referenced in SKILL.md). Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands within the context file. Capability inventory: The skill's functionality is limited to natural language generation within the agent's context. No subprocess calls, network operations, or file-write capabilities are present in the skill definition. Sanitization: There is no evidence of sanitization or filtering for the data ingested from the context file, which is used directly for stylistic alignment.
Audit Metadata