execute-spec-in-fork

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill manages internal task forks using platform-specific tools (fork_thread, read_thread, set_thread_archived). It does not invoke arbitrary shell commands or access the internet.
  • [SAFE]: Robust security logic is defined to handle incoming messages via codex-task-messenger. It explicitly instructs the agent to verify user authorization in the source thread rather than trusting content within a message card, mitigating potential spoofing or indirect injection via the messaging channel.
  • [SAFE]: The skill operates within a restricted "same-directory" scope and contains no hardcoded credentials, sensitive data exposure patterns, or external code downloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 05:14 AM
Security Audit — agent-trust-hub — execute-spec-in-fork