← 返回 Skills 市场
loop-capital

ClawOrchestrate

作者 loop-capital · GitHub ↗ · v0.1.1 · MIT-0
cross-platform ⚠ suspicious
61
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install claworchestrate
功能描述
Orchestrate AI agents across multiple machines. Dispatch tasks, monitor progress, and coordinate teams from a central gateway. Works over Tailscale, SSH, or...
使用说明 (SKILL.md)

ClawOrchestrate

Cross-machine AI agent orchestration for OpenClaw.

Dispatch tasks from one machine to agents running on another. Monitor progress. Coordinate teams.

What It Does

  • Send tasks to agents on remote machines via HTTP
  • Trigger local openclaw agent commands on remote hosts
  • Simple REST API — works with curl, any HTTP client
  • Tailscale-compatible (secure by default)

Quick Start

1. Install the dispatcher on remote machines (PC2, PC3, etc.)

# Copy dispatcher to remote machine
scp scripts/dispatcher.py user@remote-host:~/.openclaw/dispatcher.py

# Start the dispatcher
ssh user@remote-host "nohup python3 ~/.openclaw/dispatcher.py &"

2. Send tasks from your gateway machine (PC1)

# Dispatch to an agent
curl -X POST http://remote-ip:9876/dispatch \
  -H 'Content-Type: application/json' \
  -d '{"agent":"byondedu-ceo","message":"Read TASKS.md and start building."}'

# Check health
curl http://remote-ip:9876/health

3. Automate with scripts

# Dispatch to all teams
bash scripts/dispatch-all.sh "Daily standup — report progress."

Architecture

┌─────────────┐     HTTP POST      ┌─────────────┐
│   PC1       │ ─────────────────► │   PC2       │
│  (Gateway)  │                    │  (Remote)   │
│             │                    │             │
│  Che (CEO)  │    :9876/dispatch  │  ByondEdu   │
│  sends task │                    │  AgentSocial│
└─────────────┘                    └─────────────┘
                                          │
                                   dispatcher.py
                                          │
                                   openclaw agent
                                   --agent byondedu-ceo
                                   --message "..."

Setup as Systemd Service (Persistent)

On each remote machine:

cat > ~/.config/systemd/user/agent-dispatcher.service \x3C\x3C 'EOF'
[Unit]
Description=ClawOrchestrate Agent Dispatcher
After=network.target

[Service]
Type=simple
ExecStart=/usr/bin/python3 /home/%u/.openclaw/dispatcher.py
Restart=always
RestartSec=5

[Install]
WantedBy=default.target
EOF

systemctl --user daemon-reload
systemctl --user enable agent-dispatcher
systemctl --user start agent-dispatcher

API Reference

Endpoint Method Body Response
/dispatch POST {"agent":"id","message":"text"} {"status":"dispatched","agent":"id"}
/health GET — {"status":"ok"}

Security

  • Runs on Tailscale network only (not exposed to internet)
  • No authentication (Tailscale provides network-level auth)
  • Add API key auth for public deployments (see docs/SECURITY.md)

Files

  • scripts/dispatcher.py — The HTTP server (run on remote machines)
  • scripts/dispatch-all.sh — Broadcast script (run on gateway)
  • scripts/setup-service.sh — Systemd service installer
  • docs/ARCHITECTURE.md — Full architecture details
  • docs/SECURITY.md — Security hardening guide

License

MIT — Free for personal and commercial use. Premium features (dashboard, scheduler, analytics) available at claworchestrate.com.

安全使用建议
This package looks like an actual dispatcher for OpenClaw agents, but there are important inconsistencies and insecure defaults you should address before installing: - Do not deploy the systemd service as provided: the service files in SKILL.md and scripts/setup-service.sh start dispatcher without the required --api-key argument. That will cause the service to fail or, if edited incorrectly, run insecurely. - Always start dispatcher with a strong API key (use a random secret) and pass it as --api-key; do not rely on the 'changeme' default. The dispatcher enforces API key auth, but the quick-start examples omit it. - Bind address: the default bind in code is 0.0.0.0. For production limit binding to a Tailscale IP or localhost to avoid Internet exposure. Verify the bind you pass via --bind. - Environment variables: update metadata or your deployment docs to record that CLAWORCHESTRATE_KEY is used by scripts/dispatch-all.sh, and never store keys in plaintext in scripts. - Logs and data leakage: dispatcher logs message text and writes openclaw output to /tmp/dispatch-<agent>.log. If messages contain secrets, they will be written to disk. Consider rotating/secure logging or avoiding logging full messages. - Hard-coded targets and agent names: scripts/dispatch-all.sh contains specific IPs and agent IDs. Review and adapt to your own network; do not reuse those addresses or names without understanding them. - Verify OPENCLAW path: dispatcher calls a hard-coded OPENCLAW binary path (~/.local/bin/openclaw). Ensure the path is correct and that the openclaw binary cannot be abused to run arbitrary shell commands with elevated privileges. If you plan to use this skill: review the code locally, fix the systemd unit to include a secure --api-key and a safe --bind, remove insecure defaults, and test in a safe isolated network before deploying to production. If anything in the repository is unclear, ask the maintainer to explain why the documentation omits the required api-key invocation and to remove insecure defaults.
能力标签
requires-sensitive-credentials
能力评估
ℹ Purpose & Capability
Name/description (cross-machine orchestration) matches the included files: an HTTP dispatcher, broadcast script, and systemd installer. Required binaries (python3) are appropriate. However the bundle contains hard-coded private IP addresses and specific agent names (PC2/PC3, 'byondedu-ceo', etc.), which leak target topology and are unusually specific for a generic orchestration skill.
⚠ Instruction Scope
Runtime instructions are inconsistent with the code: SKILL.md Quick Start shows running dispatcher with 'python3 ~/.openclaw/dispatcher.py &' (no --api-key) but dispatcher.py requires --api-key when launched. SKILL.md also claims 'No authentication (Tailscale provides network-level auth)' while the server enforces API key auth. The systemd service templates in both SKILL.md and scripts/setup-service.sh start the dispatcher without passing an API key. The broadcast script expects/uses an Authorization header and an environment variable CLAWORCHESTRATE_KEY, but the package metadata declares no required env vars. The server logs message contents and writes agent output to /tmp logs, which could persist sensitive data. These mismatches can lead to accidental insecure deployments (exposed endpoints, default/insecure key usage, non-starting services) or operational confusion.
✓ Install Mechanism
No external install downloads; the skill is instruction+script only. That is low-risk from supply-chain/install perspective. Files are plain Python and shell scripts; nothing is fetched from arbitrary URLs or installed automatically.
⚠ Credentials
Metadata declares no required env vars, but scripts expect an API key via command-line (--api-key) and the broadcast script reads CLAWORCHESTRATE_KEY (defaulting to an insecure 'changeme'). The code writes logs to /tmp and to per-agent log files, potentially exposing message contents. There are no other unrelated credentials requested, but the lack of explicit declared env requirements and the insecure default key are disproportionate to safe operation and confusing.
ℹ Persistence & Privilege
The skill instructs installing a per-user systemd service (normal for a daemon). It does not request platform-wide privileges or 'always:true'. However the provided systemd unit files omit the required --api-key argument, which will either cause the service to fail to start or, if modified and started without a proper key, lead to a network service listening (default bind 0.0.0.0) that could be reachable beyond Tailscale. This combination increases risk if users follow instructions verbatim.
如何使用
  1. 确保已安装 OpenClaw（本地或 Docker 部署）
  2. 在对话框中输入安装命令：/install claworchestrate
  3. 安装完成后，直接呼叫该 Skill 的名称或使用 /claworchestrate 触发
  4. 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
v0.1.1
Added ClawStudio branding, product cross-links
v0.1.0
Initial release
元数据
Slug claworchestrate
版本 0.1.1
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 2
常见问题

ClawOrchestrate 是什么？

Orchestrate AI agents across multiple machines. Dispatch tasks, monitor progress, and coordinate teams from a central gateway. Works over Tailscale, SSH, or... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 61 次。

如何安装 ClawOrchestrate？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install claworchestrate」即可一键安装，无需额外配置。

ClawOrchestrate 是免费的吗？

是的，ClawOrchestrate 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。

ClawOrchestrate 支持哪些平台？

ClawOrchestrate 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 ClawOrchestrate？

由 loop-capital（@loop-capital）开发并维护，当前版本 v0.1.1。

推荐 Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191113 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 留言讨论