creative-check
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by design, as it relies on external files to define its auditing logic and processing rules.
- Ingestion points: The agent loads domain-specific rules and quality dimensions from external files including
domain.yamlandcheck-guide.md(located in$DOMAIN_DIR). It also ingests untrusted user content (creative drafts) for analysis. - Boundary markers: The instructions do not define explicit boundary markers or delimiters when interpolating rule content or draft text into the processing context, increasing the risk that embedded instructions could be obeyed by the agent.
- Capability inventory: The skill has file-system read access to the project directory and write access to generate reports, backups, and modify content files via its 'Auto-Fix' functionality.
- Sanitization: The workflow does not specify any sanitization, validation, or filtering of the content loaded from the rules or the drafts before they are used to guide the agent's behavior.
Audit Metadata