performance-report

Pass

Audited by Gen Agent Trust Hub on Oct 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external marketing data from CSV files, pastes, or connected platforms, creating an attack surface for instructions embedded in the data. * Ingestion points: Raw data provided via user input or connected platforms as described in SKILL.md. * Boundary markers: No explicit delimiters or instructions to ignore embedded prompts are present in the processing logic. * Capability inventory: The skill focuses on narrative generation and data analysis; it does not contain shell execution, file-writing, or network operation capabilities. * Sanitization: No specific validation or sanitization of input data content is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 13, 2026, 09:32 AM