orchestrate-flow

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines an internal orchestration logic for delegating work to subagents. It does not perform network requests, credential access, or arbitrary command execution.
  • [SAFE]: Operations are restricted to a specific scoped directory (.agents/temp/goals/<goal-id>/), which prevents unauthorized access to sensitive system files or other project areas.
  • [SAFE]: No obfuscation, persistence mechanisms, or privilege escalation patterns were detected. The skill uses standard markdown templates for communication between agents.
  • [SAFE]: The skill configuration explicitly disables model invocation in the frontmatter, limiting its execution environment as a defensive measure.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 01:04 PM
Security Audit — agent-trust-hub — orchestrate-flow