ecommerce-copywriting

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface. The skill is designed to ingest and process external product data provided in CSV and JSON formats for bulk description generation. 1. Ingestion points: External product data files mentioned in SKILL.md and references/product-descriptions.md. 2. Boundary markers: Absent; there are no instructions provided to the agent to treat this external data as untrusted or to ignore instructions embedded within the text fields. 3. Capability inventory: The skill is restricted to text generation and formatting for e-commerce copy. It does not have access to subprocess execution, file system modification, or network requests. 4. Sanitization: No sanitization or validation of the input data is performed. The risk is considered low because the agent's influence is limited to the content and tone of the generated copywriting.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 07:44 AM