zoom-out
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is an instructional prompt for code abstraction. It contains no executable code, network operations, or sensitive data access. The frontmatter configuration 'disable-model-invocation: true' further restricts the skill's execution environment.
- [INDIRECT_PROMPT_INJECTION]: The skill instructions include a dependency on an external local file for writing standards, creating a potential ingestion surface for untrusted instructions if that file were compromised. * Ingestion points: Local file reference to '../asd-ste100/SKILL.md' in SKILL.md. * Boundary markers: None present. * Capability inventory: The skill defines no subprocess calls, execution/eval patterns, file-write, or network operations. * Sanitization: No sanitization or validation is performed on the referenced file's content.
Audit Metadata