ce-compound-refresh
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill functions solely as a backward-compatibility layer, ensuring existing schedules and prompts continue to work while informing users of the name change. It does not perform any direct network operations or sensitive file access.
- [INDIRECT_PROMPT_INJECTION]: The skill accepts user-supplied arguments and forwards them to another skill, which represents a potential attack surface for indirect instructions.
- Ingestion points: Arguments received via the skill invocation in SKILL.md.
- Boundary markers: No explicit boundary markers or 'ignore' instructions are wrapped around the forwarded arguments.
- Capability inventory: The skill is restricted to calling the ce-compound tool.
- Sanitization: No sanitization is performed as arguments are intended to be passed unchanged to the destination skill.
Audit Metadata