afk-compass

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as a static router and recommendation engine. It reads the user's current conversation state to match it against a predefined list of specialized workflows (e.g., grill-me, wayfinder, afk-to-spec). The analysis found no evidence of malicious patterns, code execution, or network activity. The configuration disable-model-invocation: true in the skill metadata appropriately restricts the agent to follow the defined routing logic without deviating into autonomous reasoning that could bypass established guidelines. All internal cross-references to references/frontend-crossroads.md and references/workflow-crossroads.md are handled within the skill's own file structure.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 09:01 PM
Security Audit — agent-trust-hub — afk-compass