cw
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided text, such as drafts and research findings, across its various phases. It lacks explicit security boundary markers or instructions to treat this content as untrusted data. Given the skill has access to tools like Bash, WebFetch, and Write, an injection could potentially lead to unauthorized command execution or data access.
- Ingestion points: Creative writing drafts, chapters, and research data processed by the /cw-critique, /cw-voice, and /lit-review components.
- Boundary markers: Not present in the workflow or orchestration rules.
- Capability inventory: Access to Bash, WebFetch, Write, Edit, and Task tools.
- Sanitization: No evidence of input validation or escaping for external content was found.
Audit Metadata