codex-cli-yolo-task

Warn

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes the codex command-line tool with user-provided input in a shell environment.
  • [PRIVILEGE_ESCALATION]: Instructs the agent to use the --yolo or --dangerously-bypass-approvals-and-sandbox flags, which explicitly disable the security sandbox and user-approval mechanisms of the underlying tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user tasks and passes them directly to a shell-executing CLI, creating a vulnerability surface if the task content is sourced from untrusted data. * Ingestion points: User task content ingested in SKILL.md. * Boundary markers: No delimiters or specific 'ignore embedded instructions' warnings are applied to the user input before processing. * Capability inventory: Subprocess execution via codex exec in SKILL.md. * Sanitization: No validation, escaping, or filtering of the user-provided task content is performed.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 7, 2026, 12:24 PM