humanise

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted text to identify and remove AI writing patterns, creating a surface for indirect prompt injection.
  • Ingestion points: Text provided by the user for humanization (SKILL.md).
  • Boundary markers: Absent; the prompt does not use delimiters to isolate user-provided text from instructions.
  • Capability inventory: The skill utilizes file system tools (Read, Write, Edit, Grep, Glob) and user interaction (AskUserQuestion).
  • Sanitization: No sanitization or validation of the input text is mentioned or performed.
  • [EXTERNAL_DOWNLOADS]: References the Wikipedia guide for 'Signs of AI writing' and a markdown conversion service (markdown.new) in its maintenance documentation. These are well-known services used for informational purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 05:13 AM
Security Audit — agent-trust-hub — humanise