Skip to content
Back to skills

Orchestration Governance

ASecurity

Governance skill for orchestrating CrewAI specialists with strict ownership, skill boundaries, and execution permissions.

  • 6 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added May 26, 2026
ai-agentsgobashperformancedocumentation

Security analysis

A100/100

Pro scans all 5 files and shows the line behind each finding

Scanned May 27, 2026

npx -y skills add victorgrein/spec-crew --skill orchestration-governance --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Orchestration Governance?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Orchestration Governance
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/victorgrein-orchestration-governance/badge)](https://www.skillsdirectory.com/skills/victorgrein-orchestration-governance)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: orchestration-governance
version: 5.0.0
description: Governance skill for orchestrating CrewAI specialists with strict ownership, skill boundaries, and execution permissions.
---

# Orchestration Governance

## Purpose
Route user requests to the correct specialist, enforce skill boundaries, and validate outputs before returning results.

## Specialist Contracts

| Specialist | Primary purpose | Allowed skills | Permission profile |
|------------|-----------------|----------------|--------------------|
| **builder** | Build crews, agents, tasks, tools, memory | `core-build`, `tools-expert` | normal write/edit |
| **flow** | Flows, state-management, routing, orchestration, decorators | `flows` | normal write/edit |
| **auditor** | Investigation, auditing, validation | `core-build`, `flows`, `tools-expert` | read-only |
| **docs** | Documentation authoring and maintenance | `core-build`, `flows` | write/edit only `*.md`; bash read-only |

## Routing Rules

- **Builder keywords:** create, build, scaffold, setup, agent, task, crew, tool, memory
- **Flow keywords:** flow, state, router, routing, orchestration, decorator, @start, @listen, @router
- **Auditor keywords:** inspect, audit, investigate, trace, root cause, validate, performance analysis
- **Docs keywords:** docs, readme, guide, documentation, architecture notes, standards

## Canonical Command Ownership

- `/crew init` -> builder
- `/crew inspect` -> auditor
- `/crew fix` -> auditor first, then builder/flow for implementation when needed
- `/crew evolve` -> flow
- `/crew docs` -> docs

## Delegation Pattern

Write plain-language delegation with:
1. Goal
2. Context and file paths
3. Specialist skill constraints
4. Deliverables
5. Validation criteria

## Auditor Rule

Auditor executes read-only analysis and returns:
- findings
- risk assessment
- recommendations
- validation steps

Auditor does not patch files.

## Validation Checklist

- Chosen specialist matches intent and ownership
- Delegation includes exact paths and constraints
- Specialist used only allowed skills
- Permission limits are respected
- Output contract is complete and actionable

## Supporting Files

- Subagent directory: `references/subagent-directory.md`
- Delegation patterns: `references/delegation-patterns.md`
- Question templates: `references/questioning-guide.md`
- Delegation template: `assets/delegation-template.md`

Files in this skill

  • SKILL.md2.3 KB
  • assets/delegation-template.md5.1 KB
  • references/delegation-patterns.md821 B
  • references/questioning-guide.md4.7 KB
  • references/subagent-directory.md744 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.