Task tool parameters:
- subagent_type: "general-purpose"
- run_in_background: true (optional)
- prompt: |
    Consult Codex about: {topic}

    codex exec --model gpt-5.3-codex --sandbox read-only --full-auto "
    {question for Codex}
    " 2>/dev/null

    Return CONCISE summary (key recommendation + rationale).

codex exec --model gpt-5.3-codex --sandbox read-only --full-auto "Brief question" 2>/dev/null

codex exec --model gpt-5.3-codex --sandbox workspace-write --full-auto "
Implement: {task description}
Requirements: {requirements}
Files: {file paths}
" 2>/dev/null

codex exec --model gpt-5.3-codex --sandbox read-only --full-auto "
Create an implementation plan for: {feature}

Context: {relevant architecture/code}

Provide:
1. Step-by-step plan with dependencies
2. Files to create/modify
3. Key design decisions
4. Risks and mitigations
" 2>/dev/null

codex exec --model gpt-5.3-codex --sandbox read-only --full-auto "
Review this design approach for: {feature}

Context: {relevant code or architecture}

Evaluate:
1. Is this approach sound?
2. Alternative approaches?
3. Potential issues?
4. Recommendations?
" 2>/dev/null

codex exec --model gpt-5.3-codex --sandbox read-only --full-auto "
Debug this issue:

Error: {error message}
Code: {relevant code}
Context: {what was happening}

Analyze root cause and suggest fixes.
" 2>/dev/null