# Maestro Governance

> Keep Codex or ChatGPT Work tasks scoped, context-efficient, and aligned with Orquestrador Maestro routing decisions.

- Skill: `iapro-community/maestro-governance` (Agent Skill)
- Install (CLI): `npx get-skill add iapro-community/maestro-governance`
- Raw SKILL.md: https://api.skillmd.com/api/skills/iapro-community/maestro-governance/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: IAPro-Community (https://skillmd.com/u/iapro-community)
- Updated: 2026-11-11
- Page: https://skillmd.com/skills/iapro-community/maestro-governance

---


# Maestro governance

Use the Maestro hook-provided turn policy as a budget, not as a replacement for the user's request.

- Prefer the smallest useful working set of files and context.
- Load only the skill IDs selected by Maestro when those skills are available.
- Do not turn a trivial request into a Mission or multi-step orchestration flow.
- Do not spawn subagents unless the user explicitly requests parallel/multi-agent work or Maestro explicitly allows fan-out.
- Preserve Codex/Work native behavior; Maestro guidance is additive and should not force a different interface.
- If Maestro is unavailable, continue normally rather than blocking the user's task.
- Never invent token counts. Treat provider usage as unknown unless the runtime reports it.

