agent-task-splitter

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a planning and decomposition workflow for multi-agent execution environments. It specifically manages the creation of YAML plans and markdown task briefs. No code execution or network exfiltration patterns were detected in the instructions.
  • [SAFE]: The skill explicitly includes 'drift guards' and scope confirmation blocks in its templates, designed to prevent agents from exceeding their assigned file scope. This is a positive security practice for maintaining the integrity of multi-agent workflows.
  • [SAFE]: The instructions for invoking external tools (like codex or gh api) are presented as standard development workflows. Specifically, the fact-verification step using gh api targets a well-known service (GitHub) for legitimate metadata retrieval (stars, license), which falls under the trust-scope-rule for well-known services.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 12:00 PM
Security Audit — agent-trust-hub — agent-task-splitter