advanced-evaluation
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- Indirect Prompt Injection (LOW): The skill templates interpolate untrusted external data (such as
{prompt}and{response}) into its evaluation prompts. - Ingestion points: Interpolation variables in SKILL.md.
- Boundary markers: Absent in the provided templates.
- Capability inventory: None; the skill contains no scripts or tools.
- Sanitization: Not present in the markdown templates.
- Prompt Injection (SAFE): Instructional language and 'Critical' markers are used appropriately to define evaluation protocols and do not attempt to override the AI's core safety instructions.
Audit Metadata