synthesis-writing-craft
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, prompt injection, or unauthorized data access were detected. The skill follows best practices for its stated purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted prose data for editing and review purposes.
- Ingestion points: Untrusted text is ingested by the agent for craft review and editing as described in
SKILL.md. - Boundary markers: No specific boundary markers are provided for the input prose, though the skill refers to separate outcome review guidelines to verify edits.
- Capability inventory: The skill does not contain any instructions involving subprocess execution, network requests, or direct file system modifications.
- Sanitization: No explicit sanitization logic for input prose is defined in the skill files.
Audit Metadata