rank-by-user-fit

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill contains only markdown instructions and a workflow for content evaluation. There are no executable scripts, shell commands, or tools involved in the skill definition.
  • [NO_CODE]: No code or external dependencies are included. The skill functions as a pure text-based logical framework for the agent to follow when evaluating user-provided data.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external content (candidate pools). While this presents a surface for indirect prompt injection, the skill possesses no dangerous capabilities—such as file writes, network requests, or system command execution—that could be exploited if malicious content were ingested.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 02:36 PM
Security Audit — agent-trust-hub — rank-by-user-fit