linkedin-hook-extractor
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface where it ingests untrusted content from external LinkedIn posts. It proactively mitigates this by including a comprehensive 'Untrusted content' section that instructs the agent to treat fetched data strictly as data and never as instructions, specifically warning against following directions or approvals found within the fetched text.
Audit Metadata