copywriting
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user inputs (e.g., product descriptions, audience pain points) to generate marketing copy. The lack of specific delimiters or instructions to treat this data as non-executable creates a potential surface for indirect prompt injection where a malicious user could attempt to override agent instructions.
- Ingestion points: User inputs collected in the 'Gathering Requirements' section of SKILL.md (e.g., target audience, primary goal, product/service).
- Boundary markers: Absent. No specific delimiters (like XML tags or triple quotes) or 'ignore embedded instructions' warnings are provided for the interpolation of user-supplied data.
- Capability inventory: The skill is limited to text generation and possesses no capabilities for command execution, file system modification, or network requests.
- Sanitization: Absent. No evidence of input validation, escaping, or filtering before the data is processed by the model.
Audit Metadata