cw-voice-check
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from various local files to determine voice guidelines, which could contain malicious instructions intended to manipulate the agent.
- Ingestion points: The skill reads
../../references/context-contract.md, project-specificVOICE.md,STYLE.md,TASTE.md, and various content in theexamples/folder. - Boundary markers: The skill lacks explicit delimiters or specific instructions to treat the content of these files as data rather than potential commands.
- Capability inventory: The skill focuses on text diagnosis and revision; however, it does not explicitly constrain the agent's interpretation of instructions that might be embedded within the loaded voice guides.
- Sanitization: No validation or filtering is applied to the content of the external markdown files before they are incorporated into the agent's active context.
Audit Metadata