humanize
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process arbitrary prose from user-provided text or files, which creates a vulnerability surface for indirect prompt injection. Malicious instructions embedded in the input could attempt to override the agent's behavior.
- Ingestion points: The skill processes inline text and local files via the
/humanizecommand and direct natural language requests as specified inSKILL.md. - Boundary markers: The skill instructions do not explicitly require the use of delimiters (such as XML tags or triple backticks) or specific "ignore embedded instructions" warnings when handling user-provided content.
- Capability inventory: The skill is granted powerful filesystem tools including
Read,Write,Edit,Grep, andGlob, which increases the potential impact if a prompt injection attack succeeds. - Sanitization: There is no evidence of input validation, escaping, or sanitization of the external content before it is processed by the agent.
Audit Metadata