humanizer
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted data (user-provided text) and use file modification capabilities to perform its task. This creates a surface for indirect prompt injection where instructions hidden in the text to be humanized could attempt to influence the agent's behavior.
- Ingestion points: The skill accepts text directly via the
/humanizercommand or reads content from files using theReadtool. - Boundary markers: The instructions do not define specific delimiters or "ignore instructions" wrappers for the user-provided text.
- Capability inventory: The skill has access to
Write,Edit,Glob, andGrep, which allow it to modify the file system based on its processing of the input text. - Sanitization: No explicit sanitization or input filtering is performed by the skill itself; it relies on the underlying model's safety guardrails.
Audit Metadata