empatias
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed entirely of natural language instructions and reference documentation. It does not utilize any tools, perform network requests, or access the local file system beyond its own package files.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user input via responses to six guided questions. While this creates a surface for indirect prompt injection, the skill possesses no capabilities (such as code execution or data exfiltration) that could be exploited. The strictly defined output format (ranking and advice) serves as a functional constraint on the agent's behavior.
- Ingestion points: User responses to the six questions defined in
SKILL.md(Step 2). - Boundary markers: None explicitly defined for user input interpolation.
- Capability inventory: No tool usage, network operations, or file system access identified across
SKILL.mdor its referenced scripts. - Sanitization: No explicit sanitization or filtering of user input is performed before processing.
Audit Metadata