ai-evals
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Prompt Injection] (SAFE): No patterns of prompt injection or instructions to bypass safety guidelines were found in the skill body or metadata.
- [Data Exposure & Exfiltration] (SAFE): The skill does not access any files, system resources, or network endpoints. No credentials or secrets are present.
- [Obfuscation] (SAFE): All text is clear and readable. No encoded strings, zero-width characters, or homoglyphs were detected.
- [Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill does not specify any Python or Node.js dependencies, nor does it attempt to download or execute remote scripts.
- [Dynamic Execution] (SAFE): There is no evidence of runtime code generation, compilation, or unsafe deserialization.
Audit Metadata