copywriting
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided inputs to generate marketing copy and uses a script to validate file content, creating a potential surface for indirect instructions.
- Ingestion points: User input for copy generation and file paths provided to
scripts/check-output.pyvia command-line arguments. - Boundary markers: None present to distinguish between user instructions and data within the prompt or script execution.
- Capability inventory: The skill includes a local utility script (
scripts/check-output.py) that performs file reads usingpathlib.Path.read_text()to check for specific markdown headers. - Sanitization: No explicit sanitization or validation is performed on input strings or file paths before processing.
- [COMMAND_EXECUTION]: The skill includes a local Python script (
scripts/check-output.py) intended for validating agent outputs. The script is benign and only performs local file content checks without network access or privileged operations.
Audit Metadata