Skill Router

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as an orchestration layer to focus the agent on relevant tasks and prevent the unnecessary loading of broad context.
  • [DATA_EXPOSURE]: The skill references reading a local catalog.json file to match metadata such as IDs and triggers. This is restricted to the skill's own configuration data and does not involve accessing sensitive system files or user credentials.
  • [COMMAND_EXECUTION]: There are no shell commands, system calls, or script execution patterns present in the instruction set.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes user-defined goals and external catalog data, it lacks the dangerous capabilities (like network exfiltration or file modification) that would make indirect injection a viable attack vector. The ingestion of these inputs is used solely for internal routing logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 12:15 PM
Security Audit — agent-trust-hub — Skill Router