humanizer-pro
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill is designed to ingest and process 'AI drafts' which constitute untrusted external data.
- Ingestion points: Processes 'AI drafts' provided by users or external sources (implied in 'Draft Diagnostics').
- Boundary markers: There are no explicit instructions to the agent to use delimiters or ignore embedded instructions within the drafts being humanized.
- Capability inventory: The skill utilizes
activate_skillto chain multiple agent capabilities (marketing-expert, docs-pro), which could be manipulated if the input draft contains malicious instructions that override the humanization logic. - Sanitization: No sanitization or validation of input text is defined in the instructions.
Audit Metadata