humanizer
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by accepting untrusted text through humanization and detection commands. Malicious instructions embedded in the input could potentially bypass intended logic or influence agent behavior during rewriting.
- Ingestion points: The <文本> and parameters in commands like /人话, /humanize, and /detect as defined in SKILL.md.
- Boundary markers: The skill's instructions do not include the use of delimiters or specific system instructions to treat the ingested text purely as data.
- Capability inventory: The skill is restricted to text analysis and rewriting; no tools for system access, network communication, or file modification are provided.
- Sanitization: The skill documentation lacks instructions for sanitizing or escaping the content provided in user inputs before processing.
Audit Metadata