skills/andiedie/skills/ask-andie/Gen Agent Trust Hub

ask-andie

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates as a pure router, instructing the agent to evaluate current project state (work records, delivery units) and suggest the next logical step in a defined workflow.
  • [SAFE]: All referenced paths (e.g., ../and-workflow-contract/work-records.md) are relative local file references within the skill's expected environment.
  • [SAFE]: The skill explicitly defines a 'Boundary' that it is read-only, instructing the agent to name the next action and stop, without performing mutations or executing code.
  • [SAFE]: Metadata configuration (disable-model-invocation: true) and agent policy (allow_implicit_invocation: false) demonstrate a restrictive and controlled execution posture.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 06:01 AM
Security Audit — agent-trust-hub — ask-andie