editorial-copy
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest a user-provided 'brief' to guide the content generation process, which represents a potential surface for indirect prompt injection.\n
- Ingestion points: The brief is processed in
SKILL.mdduring Step 1 to identify the editorial sub-type.\n - Boundary markers: The instructions do not define explicit delimiters or 'ignore' directives for the content of the brief, relying instead on the agent's internal logic.\n
- Capability inventory: The skill's capabilities are strictly confined to text generation and scoring; it lacks any tools for file system modification, network requests, or shell command execution.\n
- Sanitization: No sanitization is performed on the input, but the mandatory 'De-slop pass' (Step 5) and 'Quality rubric' (Step 6) act as effective content filters before final output.
Audit Metadata