← 返回 Skills 市场
multi-agents-orchestration
作者
Chaojie Chen
· GitHub ↗
· v1.0.0
· MIT-0
272
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install multi-agents-orchestration
功能描述
定义多 Agent 团队两种核心协作模式：后台 spawn 实现统一调度与汇总，Discord @mention 实现角色外显与公开协作。
安全使用建议
This skill is a documentation/configuration guide — it does what it says. Before applying examples: (1) avoid storing Discord BOT_TOKEN or other secrets in world-readable files; prefer secure secret storage and least-privilege bot tokens (narrow scopes, limited guilds/channels). (2) Don't use overly-broad allowances like subagents.allowAgents: ["*"] unless you understand the risk; restrict which agents can be spawned. (3) When exposing agents to Discord, review channel permissions and rate limits to prevent accidental public disclosure or bot loops (Bot-Bot @mention). (4) Test changes in a safe environment before applying to production gateway instances. If you need tighter assurance, ask the skill author for provenance or an explicit change log — this file is instructional only and contains no executable code.
功能分析
Type: OpenClaw Skill
Name: multi-agents-orchestration
Version: 1.0.0
The skill bundle provides a comprehensive methodology and configuration guide for multi-agent orchestration within the OpenClaw ecosystem. It details legitimate orchestration patterns using the sessions_spawn function and Discord integrations, providing templates for configuration files like openclaw.json and SOUL.md. No malicious intent, data exfiltration, or harmful instructions were found in SKILL.md or _meta.json; the content is purely educational and operational for its intended framework.
能力评估
Purpose & Capability
The name/description match the SKILL.md content: it documents multi‑agent orchestration modes, configuration files, and operational steps. All requested concepts (spawn, discord binding, workspaces) are relevant to the stated purpose.
Instruction Scope
The instructions stay within the stated scope: they show how to register agents, spawn subagents, restart the gateway, and optionally expose agents as Discord bots. The SKILL.md references local OpenClaw config files and example Discord token fields only as part of the configuration process and does not instruct the agent to read unrelated system secrets or exfiltrate data.
Install Mechanism
There is no install spec and no code files. The skill is instruction-only, so it does not write code to disk or download external packages.
Credentials
The skill itself declares no required env vars, but the examples show storing Discord BOT_TOKEN and configuring workspaces/openclaw.json. Supplying Discord tokens or using wildcard subagents (e.g., subagents.allowAgents: ["*"]) is expected for Discord integration but increases privilege surface and should be limited and stored securely.
Persistence & Privilege
always:false and no install means the skill does not force permanent inclusion or modify other skills' configs. It instructs the user to restart the OpenClaw gateway for config changes, which is normal operational behavior.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install multi-agents-orchestration - 安装完成后，直接呼叫该 Skill 的名称或使用
/multi-agents-orchestration触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
OpenClaw 多 Agent 团队编排方法论。定义了后台调度（Backend Spawn）与 Discord @Mention 两种协作模式的本质区别、如何共存并行、三种部署模式选择（纯后台 / 混合 / 全外显）、Agent vs Skill 决策框架、完整搭建流程、Discord 三层覆盖机制、以及常见故障排查。适用于需要搭建或调整 OpenClaw 多 Agent 团队协作的场景。
元数据
常见问题
multi-agents-orchestration 是什么？
定义多 Agent 团队两种核心协作模式：后台 spawn 实现统一调度与汇总，Discord @mention 实现角色外显与公开协作。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 272 次。
如何安装 multi-agents-orchestration？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install multi-agents-orchestration」即可一键安装，无需额外配置。
multi-agents-orchestration 是免费的吗？
是的，multi-agents-orchestration 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
multi-agents-orchestration 支持哪些平台？
multi-agents-orchestration 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 multi-agents-orchestration？
由 Chaojie Chen（@cchaojiejes）开发并维护，当前版本 v1.0.0。
推荐 Skills