persuasive-argument
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is purely instructional, focusing on providing a framework for persuasive writing. It does not execute scripts, download external content, or perform network operations outside of standard platform capabilities.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data as the basis for its writing tasks, which creates a standard attack surface for indirect prompt injection.
- Ingestion points: The skill explicitly requests user input for the 'claim', 'audience' details, and 'evidence'.
- Boundary markers: The skill uses Markdown headers and code blocks to structure its output, though it does not implement explicit 'ignore' delimiters for the user-supplied content.
- Capability inventory: The skill's actions are limited to text generation and an optional call to a user-authorized notes connector (e.g., Scratchpad connector).
- Sanitization: No specific input sanitization or filtering is defined within the skill instructions, relying on the underlying model's safety guardrails.
Audit Metadata