copywriting-expert
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is composed entirely of reference documentation and instructions without any associated scripts or executable files, eliminating risks associated with malicious code execution or privilege escalation within the skill itself.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and process untrusted data from the local project environment, such as README files, UI strings, and documentation, which could potentially contain instructions meant to bias the agent's output.
- Ingestion points: The agent is directed to read existing project files, code, and READMEs in SKILL.md and references/project-source-of-truth.md.
- Boundary markers: No specific delimiters are provided for untrusted data, though the skill emphasizes project-specific truth.
- Capability inventory: The skill has no scripts and relies on the platform's native tools for file reading; it does not perform network writes or file modifications.
- Sanitization: None defined, as the skill produces text suggestions for human review.
Audit Metadata