copywritting
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted user data (such as emails, ad copy, and landing page text) to perform rewriting and polishing tasks. This creates a surface for indirect prompt injection where malicious instructions could be embedded in the text provided by a user.
- Ingestion points: The skill accepts arbitrary text inputs for tasks defined as 'polish', 'rewrite', or 'create' in
SKILL.md. - Boundary markers: The instructions provide conceptual boundaries by requiring the agent to classify the request type and 'Protect the mandate', although no technical delimiters (like XML tags) are enforced for the input data.
- Capability inventory: The skill is strictly limited to text generation. There are no tools or scripts provided that allow for file system writes, network operations, or shell command execution.
- Sanitization: There is no evidence of input sanitization or filtering to detect instructions embedded within the text to be rewritten.
Audit Metadata