conversion-copy
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from a user-provided 'brief' to generate copy drafts.
- Ingestion points:
SKILL.mdworkflow steps 1 and 4 ingest data from the 'brief' field to identify copy types and draft content. - Boundary markers: No specific boundary markers or 'ignore' instructions are used to separate user data from the system prompt.
- Capability inventory: The skill lacks access to any sensitive tools, such as network operations, filesystem access, or command execution.
- Sanitization: No input sanitization or validation of the brief content is performed.
Audit Metadata