voice-reviewer
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external content files (Markdown, MDX), creating a surface for indirect prompt injection where instructions embedded in the data could influence agent behavior.
- Ingestion points: The agent reads files from the project directory, including 'docs/brand-voice-guidelines.md', '.voice-lint.json', and the specific content files targeted for review.
- Boundary markers: The instructions lack explicit boundary markers or warnings to the agent to ignore instructions contained within the content files being audited.
- Capability inventory: The agent has the capability to read local project files and generate suggestions/reports based on the content analyzed.
- Sanitization: No mechanisms for sanitizing or filtering the ingested content are defined in the workflow.
Audit Metadata