Menu
Pre-development audit checklist for product packages before build. Reference material (non-actionable knowledge).
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
| name | pre-dev-audit |
| description | Pre-development audit checklist for product packages before build. Reference material (non-actionable knowledge). |
Load and follow the instructions from the canonical file at .vaibe/skills/pre-dev-audit/SKILL.md
as if they were written here. Do not act on this wrapper — it holds only metadata; the full
skill logic lives in the canon.
@.vaibe/skills/pre-dev-audit/SKILL.md
Baseado na classificação ocupacional SOC
Mandatory prerequisite for any skill work — load this FIRST and build from it whenever authoring, reviewing, or changing a skill, instead of pattern-completing from another skill as a template. "Reference" names its content type, not its priority: it is not optional. The sourced reference for how agent skills are composed — the cross-tool open standard (agentskills.io) and the deltas for Cursor, Claude Code, OpenCode, and Codex, plus Anthropic authoring best practices, and how it maps onto the vAIbe-OS canon (model A — .vaibe/ canon + generated native layer). Triggers: skill authoring, creating a skill, editing a skill, SKILL.md, frontmatter, native layer, agentskills.
Mandatory prerequisite for any skill work — load this FIRST and build from it whenever authoring, reviewing, or changing a skill, instead of pattern-completing from another skill as a template. "Reference" names its content type, not its priority: it is not optional. The sourced reference for how agent skills are composed — the cross-tool open standard (agentskills.io) and the deltas for Cursor, Claude Code, OpenCode, and Codex, plus Anthropic authoring best practices, and how it maps onto the vAIbe-OS canon (model A — .vaibe/ canon + generated native layer). Triggers: skill authoring, creating a skill, editing a skill, SKILL.md, frontmatter, native layer, agentskills.
Mandatory prerequisite for any skill work — load this FIRST and build from it whenever authoring, reviewing, or changing a skill, instead of pattern-completing from another skill as a template. "Reference" names its content type, not its priority: it is not optional. The sourced reference for how agent skills are composed — the cross-tool open standard (agentskills.io) and the deltas for Cursor, Claude Code, OpenCode, and Codex, plus Anthropic authoring best practices, and how it maps onto the vAIbe-OS canon (model A — .vaibe/ canon + generated native layer). Triggers: skill authoring, creating a skill, editing a skill, SKILL.md, frontmatter, native layer, agentskills.
Mandatory prerequisite for any skill work — load this FIRST and build from it whenever authoring, reviewing, or changing a skill, instead of pattern-completing from another skill as a template. "Reference" names its content type, not its priority: it is not optional. The sourced reference for how agent skills are composed — the cross-tool open standard (agentskills.io) and the deltas for Cursor, Claude Code, OpenCode, and Codex, plus Anthropic authoring best practices, and how it maps onto the vAIbe-OS canon (model A — .vaibe/ canon + generated native layer). Triggers: skill authoring, creating a skill, editing a skill, SKILL.md, frontmatter, native layer, agentskills.
Mandatory prerequisite for any skill work — load this FIRST and build from it whenever authoring, reviewing, or changing a skill, instead of pattern-completing from another skill as a template. "Reference" names its content type, not its priority: it is not optional. The sourced reference for how agent skills are composed — the cross-tool open standard (agentskills.io) and the deltas for Cursor, Claude Code, OpenCode, and Codex, plus Anthropic authoring best practices, and how it maps onto the vAIbe-OS canon (model A — .vaibe/ canon + generated native layer). Triggers: skill authoring, creating a skill, editing a skill, SKILL.md, frontmatter, native layer, agentskills.
Create or update project management plan with task statuses and priorities. Triggers: plan, sprint plan, plan-update, обновить план, план проекта, точка сборки, личный план, план работ, сохрани план, сохрани в планы.
