codex-task-to-spec

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes a local Node.js script (scripts/review-gate.mjs) to validate workflow checkpoints and reviewer reports. This execution is confined to the local filesystem and uses only built-in modules.\n- [PROMPT_INJECTION]: The skill handles natural language tasks by isolating them in a dedicated 'authority' log and using a multi-stage architecture to prevent sub-agents from executing instructions found within project evidence files.\n- [SAFE]: All operations are restricted to the local development environment. The skill does not perform network requests, access sensitive system credentials, or use obfuscated code.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 10:59 AM
Security Audit — agent-trust-hub — codex-task-to-spec