skill-topic-evaluator
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security threats or malicious patterns were detected in the skill instructions or metadata files. The skill functions purely as a prompt template for evaluation logic.- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data via the topic description and background parameters in SKILL.md. However, it lacks exploitable capabilities such as file system writing, command execution, or network access. There are no explicit boundary markers or sanitization routines mentioned, but the risk remains low due to the limited capability profile.- [DATA_EXFILTRATION]: The skill references local profile files (identity.md, audience.md, etc.) and shared configuration (scoring-dimensions.md) using relative paths. This is used for context-aware scoring and does not involve accessing sensitive system directories (like .ssh or .aws) or exfiltrating data to external servers.
Audit Metadata