copy-editing
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues were identified in the skill. The files consist of instructional markdown and evaluation data for a copy-editing persona.
- [DATA_EXPOSURE]: The skill instructs the agent to read local project context files (e.g.,
.agents/product-marketing.mdor.claude/product-marketing.md) if they exist. These are standard platform-specific context files used to guide the agent's brand voice and do not represent the exposure of sensitive system credentials or private keys. - [INDIRECT_PROMPT_INJECTION]: The skill acts on user-provided text, which introduces an indirect prompt injection surface. However, the risk is negligible as the skill does not possess any capabilities (such as network access, file writing, or command execution) that could be exploited by malicious text.
- Ingestion points: User-provided marketing copy for editing, as described in
SKILL.md. - Boundary markers: The instructions do not define explicit boundary markers for user input.
- Capability inventory: No use of
eval(),exec(), subprocess calls, file-write operations, or network requests across any files. - Sanitization: No specific sanitization or filtering of input text is implemented.
Audit Metadata