dag-planner

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a planning framework for task decomposition. It does not include executable code, scripts, or binary files.
  • [DATA_EXPOSURE]: No patterns for accessing sensitive files (e.g., credentials, SSH keys) or environment variables were detected. The skill uses standard file-system tools (Read, Write, Edit) solely for managing DAG configuration files.
  • [REMOTE_CODE_EXECUTION]: There are no external downloads, package installations (npm/pip), or remote script execution patterns (curl|bash).
  • [PROMPT_INJECTION]: The instructions do not attempt to bypass agent safety filters or override system-level constraints. The 'IMPORTANT' and 'NOT-FOR' sections correctly define operational boundaries for the agent.
  • [INDIRECT_PROMPT_INJECTION]: While the skill ingests untrusted user task descriptions to generate DAG nodes, it is designed for structural planning. The instructions emphasize explicit input/output schemas and validation, which serve as foundational safety controls for downstream execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 07:19 PM
Security Audit — agent-trust-hub — dag-planner