cw-reader
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external text (copy) which may contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The skill reads user-provided copy for review and accesses external project context from
../../references/context-contract.mdandAUDIENCE.md. - Boundary markers: The instructions lack explicit boundary markers or instructions for the agent to ignore potentially malicious commands embedded within the text being reviewed.
- Capability inventory: While the skill itself does not request network or filesystem-write tools, it operates within an environment where the agent likely has the ability to read project files and generate complex outputs based on untrusted data.
- Sanitization: There are no sanitization steps defined to filter or escape instructions embedded in the provided copy.
Audit Metadata