style-writer
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates strictly within its intended domain of text analysis and generation. All file system activities, such as reading templates and saving user-defined styles, are restricted to the skill's local directories.
- [INDIRECT_PROMPT_INJECTION]: The skill handles user-provided text to extract writing styles, creating a standard data ingestion surface.
- Ingestion points: User text provided via Input Mode C and Path 3 as described in SKILL.md.
- Boundary markers: The EXTRACTION_ENGINE.md defines a comprehensive 'Pollution Check' designed to filter out content-specific elements like names and locations to isolate stylistic metadata.
- Capability inventory: The skill is limited to guiding LLM generation and writing style profiles to the local 'my-styles' directory; it lacks dangerous capabilities like arbitrary command execution or network access.
- Sanitization: The extraction process converts input text into a structured JSON schema of style parameters, which effectively decouples user input from the generation engine's control flow.
Audit Metadata