style-writer
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external, potentially untrusted text including voice profiles (
voice-dna.md) and source materials for factual claims. - Ingestion points: The agent is instructed to read local Markdown files (
voice-dna.md,style-dna.md) and user-provided writing briefs/source materials. - Boundary markers: The skill does not define explicit delimiters or 'ignore' instructions for the source content it processes.
- Capability inventory: The skill is limited to text generation and stylistic measurement. It does not contain instructions for subprocess execution, network requests, or arbitrary file system writes.
- Sanitization: No explicit sanitization or filtering of the input text is performed, though the skill contains internal directives to avoid fabricating facts or endorsements based on the input style.
Audit Metadata