linkedin-roast
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from LinkedIn profiles via the
get_page_texttool andcomputerscreenshots as described in the workflow. 1. Ingestion points: Untrusted content is ingested from LinkedIn profile pages via theget_page_texttool inSKILL.md. 2. Boundary markers: The skill uses a structured roast template and explicit comedy rules to guide the agent. 3. Capability inventory: The skill utilizes text extraction and screenshot capabilities to analyze and roast profiles. 4. Sanitization: The skill includes 'Golden Rules' that prohibit offensive content and protected characteristics, reducing the impact of potentially malicious input.
Audit Metadata