oz-child-agent-orchestration

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the use of an environment-provided CLI tool, referenced via the OZ_CLI variable, to execute messaging commands and manage orchestration tasks.
  • [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface by instructing the agent to read and act upon message content from a parent agent.
  • Ingestion points: Task instructions are received through the "$OZ_CLI" run message read command in SKILL.md.
  • Boundary markers: No specific boundary delimiters or safety warnings are provided for message content.
  • Capability inventory: The skill enables the agent to execute shell commands and create file patches based on subtask descriptions.
  • Sanitization: The skill does not describe any validation or sanitization of the data retrieved from the messaging system.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 04:59 AM
Security Audit — agent-trust-hub — oz-child-agent-orchestration