linkedin-roast
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- PROMPT_INJECTION (LOW): The skill possesses an indirect prompt injection surface due to its reliance on external LinkedIn profile content. * Ingestion points: Profile data is collected using
get_page_textand thecomputerscreenshot tool as described inSKILL.md. * Boundary markers: Absent; there are no delimiters or instructions provided to separate external profile data from the agent's roasting instructions. * Capability inventory: The skill's capabilities are limited to text generation and reading; no scripts, network operations, or file system modifications are present in the skill files. * Sanitization: No filtering or sanitization of the input text is implemented before processing. - NO_CODE (SAFE): The skill consists entirely of markdown templates and documentation with no executable scripts or binaries provided, eliminating risk from direct code execution.
Audit Metadata