persona-design

Pass

Audited by Gen Agent Trust Hub on May 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely instructional and consists of Markdown text. It provides a methodology for defining AI roles, capability boundaries, and behavioral styles without any functional code or system interactions.
  • [PROMPT_INJECTION]: While the text mentions 'leaked prompts' in a historical and analytical context, it does not contain any instructions that attempt to bypass safety filters, extract system prompts, or override the agent's core safety guidelines.
  • [DATA_EXFILTRATION]: There are no commands or instructions that access sensitive files, environment variables, or perform network operations to send data externally.
  • [REMOTE_CODE_EXECUTION]: The skill does not include any scripts, package installations, or remote download commands (e.g., curl, wget).
  • [OBFUSCATION]: A review of the file found no evidence of Base64 encoding, hex-encoded strings, zero-width characters, or hidden Unicode tags.
Audit Metadata
Risk Level
SAFE
Analyzed
May 4, 2026, 01:59 PM