sepia-refactor
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a wrapper for a specific operation (refactor) located in a sibling directory (../sepia/SKILL.md). This is a standard pattern for modular skill distribution.
- [INDIRECT_PROMPT_INJECTION]: The instructions explicitly direct the agent to treat target content as untrusted data and not as instructions or authority. This is a recommended safety practice to mitigate indirect prompt injection risks when processing user-provided prose.
- [COMMAND_EXECUTION]: The skill instructions explicitly state that invoking the entry grants no tool, file, network, or external-action authority, and no dangerous commands are present.
Audit Metadata