Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Multi Agent Orchestrator

ASecurity

当有 3 个以上专职 AI Agent 需协同完成复杂任务,且出现重复劳动、冲突改文件或质量参差时使用;做的事是搭建单一编排器(任务拆解→去重→路由→质量门禁→心跳监控)并产出任务登记表、委派指令与验收证据;不适用于单 Agent 简单任务或无需协同的场景;触发词:多智能体编排、multi-agent orchestration、任务路由 task routing、Agent 去重 anti-duplication、质量门禁 quality gate、心跳监控 heartbeat

3 stars
0 votes
0 copies
1 views
Added 9/19/2026
ai-agentspythonsqlcode-reviewgitapisecurity

Works with

cursorcliapi

Security Analysis

A100/100

Scanned 9/19/2026

$npx -y skills add findscripter/everything-skills --skill multi-agent-orchestrator --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Multi Agent Orchestrator?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Multi Agent Orchestrator
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/findscripter-multi-agent-orchestrator/badge)](https://www.skillsdirectory.com/skills/findscripter-multi-agent-orchestrator)

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

Download with Pro
Files
SKILL.md
---
name: multi-agent-orchestrator
title: 多智能体任务编排器
description: 当有 3 个以上专职 AI Agent 需协同完成复杂任务,且出现重复劳动、冲突改文件或质量参差时使用;做的事是搭建单一编排器(任务拆解→去重→路由→质量门禁→心跳监控)并产出任务登记表、委派指令与验收证据;不适用于单 Agent 简单任务或无需协同的场景;触发词:多智能体编排、multi-agent orchestration、任务路由 task routing、Agent 去重 anti-duplication、质量门禁 quality gate、心跳监控 heartbeat
domain: 协作/automation
triggers: [多智能体编排, multi-agent orchestration, 任务路由, task routing, Agent 去重, anti-duplication, 质量门禁, quality gate, 心跳监控, heartbeat]
tags: [multi-agent, orchestration, task-routing, quality-gates, anti-duplication]
level: 进阶
status: stable
agents: [claude-code, codex, cursor, gemini-cli]
tools: [python, sqlite3, difflib, git, claude, cursor, gemini]
requires: []
related: [agent-workflow-builder, task-decomposition-planner, multi-agent-system-designer, parallel-agent-hub]
combines_with: [task-decomposition-planner, agent-workflow-builder, multi-agent-workflow-designer]
license: MIT
source: sickn33/agentic-awesome-skills
source_license: MIT
---
## 何时使用

适用场景:

- 有 3 个及以上专职 Agent(代码、研究、测试、安全、文档等)需要在一个复杂任务上协同。
- Agent 之间出现重复劳动或互相冲突(例如同时改同一文件)。
- 需要审计轨迹,记录"谁、在何时、做了什么"。
- Agent 产出质量不稳定,需要在标记完成前做验证。

不该用:

- 单个 Agent 即可完成、无需拆解与协同的简单任务。
- 不要把编排器的产出当作环境相关验证、测试或专家评审的替代品。
- 若缺少必要输入、权限、安全边界或验收标准,应先停下来澄清,而不是开工。

## 步骤

1. 定义编排器身份(含 NOT-block):明确它"是什么 / 不是什么",强制它只委派不亲自干活。生产数据显示 NOT-block 可降低约 35% 的任务漂移。
2. 建立任务登记表:在分配前用相似度比对查重,避免重复派单。
3. 路由到专职 Agent:用关键词打分把任务匹配到最合适的 Agent。
4. 执行质量门禁:把 Agent 自报"完成"视为"主张",把测试输出视为"证据",全部检查通过才标记 done。
5. 每 30 分钟心跳:盘点近 30 分钟委派了什么,发现空闲 Agent 就补派或重派。

## 指令

身份定义(NOT-block 模式):

```
你是任务编排器(Task Orchestrator)。你绝不亲自做专项工作。
你负责:拆解任务、委派给正确的 Agent、防止冲突、并在标记完成前验证质量。

你不是:
- 不是写代码的人 —— 委派给代码 Agent
- 不是研究员 —— 委派给研究 Agent
- 不是测试员 —— 委派给测试 Agent
```

任务查重(SQLite + difflib,相似度阈值 0.55):

```python
import sqlite3
from difflib import SequenceMatcher

def check_duplicate(description, threshold=0.55):
    conn = sqlite3.connect("task_registry.db")
    c = conn.cursor()
    c.execute("SELECT id, description, agent, status FROM tasks WHERE status IN ('pending', 'in_progress')")
    for row in c.fetchall():
        ratio = SequenceMatcher(None, description.lower(), row[1].lower()).ratio()
        if ratio >= threshold:
            return {"id": row[0], "description": row[1], "agent": row[2]}
    return None
```

关键词打分路由:

```python
AGENTS = {
    "code-architect": ["code", "implement", "function", "bug", "fix", "refactor", "api"],
    "security-reviewer": ["security", "vulnerability", "audit", "cve", "injection"],
    "researcher": ["research", "compare", "analyze", "benchmark", "evaluate"],
    "doc-writer": ["document", "readme", "explain", "tutorial", "guide"],
    "test-engineer": ["test", "coverage", "unittest", "pytest", "spec"],
}

def route_task(description):
    scores = {}
    for agent, keywords in AGENTS.items():
        scores[agent] = sum(1 for kw in keywords if kw in description.lower())
    return max(scores, key=scores.get) if max(scores.values()) > 0 else "code-architect"
```

质量门禁(全部通过才算 done):

```
Agent 报告完成后逐项核验:
1. 文件是否真的改动了?  git diff --stat
2. 测试是否通过?        npm test / pytest
3. 是否引入了密钥?      grep 检索 API key、token
4. 构建是否成功?        npm run build
5. 是否只动了该动的文件?(范围检查)
仅当全部检查通过,才标记完成。
```

30 分钟心跳:

```
每 30 分钟自问:
1. 过去 30 分钟我"委派"了什么?
2. 若什么都没委派 → 打开任务积压,派下一个任务
3. 检查空闲 Agent(分配的任务上 >30 分钟无消息)
4. 重新催促空闲 Agent,或重新分配其任务
```

## 示例

示例 1 —— 委派一个代码任务(明确范围、验证命令、截止时间):

```
[ORCHESTRATOR -> code-architect] 任务:给 /api/users 加限流
范围:仅 src/middleware/rate-limit.ts
验证:npm test -- --grep "rate-limit"
截止:30 分钟
```

示例 2 —— 处理重复任务:

```
用户提出:"修复登录 bug"
登记表查重:任务 #47 "Fix authentication bug" 正由 security-reviewer 进行中
决策:SKIP —— 已有相似任务(78% 匹配)
动作:告知用户已有任务,等待其完成
```

## 注意事项

- 给每个 Agent 都写 NOT-block,明确它必须拒绝做什么。
- 任务登记表用 SQLite 即可(轻量、无需服务端)。
- 去重相似度阈值取 55%(再低会误报过多);可对同名文件做行级锁与队列,避免两个 Agent 同时改一个文件。
- 质量门禁要基于"证据"而非 Agent 的口头主张:先看 git diff 再接受完成。
- 每次委派都留痕:任务 ID、Agent、范围、截止时间、验证命令。
- 常见坑:编排器自己动手做活(对策:加 NOT-block 与角色边界);任务堆积无进展(对策:靠 30 分钟心跳发现并重派陈旧任务)。

## 互见

- code-reviewer:委派后对代码改动做评审。
- dependency-auditor:在质量门禁中核查依赖与安全问题。

---

本条采编自 sickn33/antigravity-awesome-skills(MIT)。

Attribution

findscripterfindscripter
View sourceSee grades on GitHubMore from findscripter →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

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 (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698621 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →