copywriting
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of Markdown instructions and reference material designed to guide an AI agent in writing marketing copy. There are no instances of command execution, network requests to external domains, or hardcoded credentials.
- [NO_CODE]: No executable scripts (Python, JavaScript, or Shell) are present in the skill package. All functionality is delivered through natural language prompting and structured references.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a data ingestion surface by reading local context files such as
.agents/product-marketing.md. 1. Ingestion points: SKILL.md specifies reading product marketing context from local files. 2. Boundary markers: The skill does not explicitly define delimiters for this external content. 3. Capability inventory: The skill has no capabilities to execute code, write files, or access the network. 4. Sanitization: No sanitization is performed on the ingested content. Despite the lack of sanitization and markers, the severity is SAFE because the skill lacks the tools necessary to perform any harmful actions if an injection were to occur.
Audit Metadata