prose-writing

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill utilizes standard instructional techniques for creative writing and references a project-specific directory for configuration files.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external briefs and style files, which are potential ingestion points for third-party instructions. However, the risk is minimal as no high-privilege tools are enabled.
  • Ingestion points: Briefs provided in the prompt and style files located in "$MERIDIAN_FS_DIR/styles/".
  • Boundary markers: Absent.
  • Capability inventory: The skill defines behavioral instructions for text generation and does not utilize external tools.
  • Sanitization: None provided for input data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 01:14 AM