codex-builder

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill documentation instructs the user to run a local script skills/codex-builder/scripts/execute-plan.sh which executes commands to implement code via the Codex CLI.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its ingestion of external plan files.\n
  • Ingestion points: Markdown plan files located in ~/.claude/plans/ or /tmp/.\n
  • Boundary markers: None identified; the skill directly processes the file content.\n
  • Capability inventory: Includes file modification and creation across the repository via Codex CLI delegation and local script execution.\n
  • Sanitization: No validation or sanitization of the plan file content is mentioned in the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 11:33 PM