voice-check
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content (a 'substacker draft') as its primary data source. While the implementation is focused on text analysis and regex matching, processing arbitrary user text can serve as a vector for indirect prompt injection if the draft contains hidden instructions designed to influence the agent.
- Ingestion points: The skill ingests and scans text from a 'substacker draft' as defined in the workflow of SKILL.md.
- Boundary markers: No explicit delimiters (e.g., XML tags or triple backticks) or 'ignore embedded instructions' warnings are used when processing the draft content.
- Capability inventory: The skill's capabilities are limited to reading local configuration files (voice-profile.md) and emitting a formatted table. No network access, file system modifications, or shell execution tools are used.
- Sanitization: No input validation or sanitization of the draft content is performed before processing.
Audit Metadata