ethical-persuasion-design
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely composed of Markdown documentation and JSON evaluation files. No scripts, executables, or automated tool calls are present.
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to audit and rewrite user-provided marketing copy, which serves as an ingestion point for untrusted data. However, the skill lacks any capabilities (such as file system access, network tools, or shell execution) that could be abused if an injection were successful.
- Ingestion points: User-provided copy, launch offers, and behavior-change messages processed in
workflows/design-persuasive-request.md. - Boundary markers: None; the instructions do not explicitly require the use of delimiters to separate user input from agent instructions.
- Capability inventory: None; no subprocesses, file operations, or network calls are defined in any file.
- Sanitization: None detected.
Audit Metadata