human-sounding-copy

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists entirely of instructional Markdown files intended to guide the AI agent's writing style. No security issues were detected.
  • [NO_CODE]: There are no executable scripts (e.g., .py, .js, .sh) or binary files included in this skill. The skill operates purely through natural language instructions.
  • [PROMPT_INJECTION]: The instructions are focused on text transformation and style matching. There are no attempts to override safety filters, bypass instructions, or manipulate the agent's core operational logic.
  • [DATA_EXFILTRATION]: The skill does not access sensitive files, environment variables, or credentials. It performs no network operations and has no mechanisms for sending data to external servers.
  • [EXTERNAL_DOWNLOADS]: While several reference files contain URLs to well-known blogs and research papers (e.g., paulgraham.com, stripe.com), these are used for informational purposes and style grounding. There are no automated downloads or remote code execution patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 03:12 PM
Security Audit — agent-trust-hub — human-sounding-copy