humanize
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external text which could potentially contain malicious instructions or prompt injection attempts.
- Ingestion points: The skill accepts text input via direct user prompts, file content specified by path, and embedded calls from other AI agent skills.
- Boundary markers: The instruction set in SKILL.md includes a dedicated security section that explicitly directs the agent to treat input as content only and ignore any embedded imperatives or commands.
- Capability inventory: The skill has the ability to read and write to the local filesystem to update documents with humanized versions of the text.
- Sanitization: The skill mandates that detected instructions within input text are either edited as prose or reported to the user, preventing unauthorized command execution.
Audit Metadata