Agent of Empires Session Manager
agent-of-empires/agent-of-empires
Starts, monitors and organizes coding agent sessions that run in tmux through the aoe command, including groups, profiles and worktree-based parallel work.
Orchestrix multi-agent workflow guide for OpenClaw. An agent skill from LeoYeAI/openclaw-master-skills.
$ npx skills add LeoYeAI/openclaw-master-skills --skill orchestrix-guide -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install LeoYeAI/openclaw-master-skills orchestrix-guide --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/orchestrix-guide .claude/skills/orchestrix-guide && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "orchestrix-guide" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/orchestrix-guide into .claude/skills/orchestrix-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "orchestrix-guide", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/orchestrix-guideType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add LeoYeAI/openclaw-master-skills --skill orchestrix-guide -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install LeoYeAI/openclaw-master-skills orchestrix-guide --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/orchestrix-guide .agents/skills/orchestrix-guide && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "orchestrix-guide" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/orchestrix-guide into .agents/skills/orchestrix-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "orchestrix-guide", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add LeoYeAI/openclaw-master-skills --skill orchestrix-guide -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install LeoYeAI/openclaw-master-skills orchestrix-guide --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/orchestrix-guide .cursor/skills/orchestrix-guide && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "orchestrix-guide" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/orchestrix-guide into .cursor/skills/orchestrix-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "orchestrix-guide", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/LeoYeAI/openclaw-master-skills.git --path skills/orchestrix-guide--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add LeoYeAI/openclaw-master-skills --skill orchestrix-guide -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install LeoYeAI/openclaw-master-skills orchestrix-guide --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/orchestrix-guide .gemini/skills/orchestrix-guide && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "orchestrix-guide" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/orchestrix-guide into .gemini/skills/orchestrix-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "orchestrix-guide", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install LeoYeAI/openclaw-master-skills orchestrix-guideInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add LeoYeAI/openclaw-master-skills --skill orchestrix-guide -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/orchestrix-guide .github/skills/orchestrix-guide && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "orchestrix-guide" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/orchestrix-guide into .github/skills/orchestrix-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "orchestrix-guide", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add LeoYeAI/openclaw-master-skills --skill orchestrix-guide -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install LeoYeAI/openclaw-master-skills orchestrix-guide --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/orchestrix-guide .opencode/skills/orchestrix-guide && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "orchestrix-guide" agent skill from https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/orchestrix-guide into .opencode/skills/orchestrix-guide/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "orchestrix-guide", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
orchestrix-guideOrchestrix multi-agent workflow guide for OpenClaw. An agent skill from LeoYeAI/openclaw-master-skills.
Orchestrix Guide is an agent skill from LeoYeAI/openclaw-master-skills. Orchestrix multi-agent workflow guide for OpenClaw. Defines two operational phases: (1) Planning Phase — sequential agent orchestration in a single tmux window from project-brief through PRD, UX spec, architecture, to PO shard; (2) Development Phase — automated multi-window tmux collaboration via HANDOFF. Includes tmux send-keys protocol, task completion detection, and supplementary flows (bug fix, iteration, brownfield, change management).
Its SKILL.md is about 4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `_meta.json`).
It sits in Agent Workflows, covering PRD writing, Multi-agent orchestration and Debugging. It works with tmux. The repository describes itself as: 🧠 Curated collection of 1209+ best OpenClaw skills — weekly updated by MyClaw.ai. The licence is MIT.
4 steps, taken from the first numbered list in SKILL.md.
Read from SKILL.md and the folder at commit e5199b5. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Shell commands in SKILL.md call:
bashclaudebrewgitFrom the folder's file list and the shell code blocks in SKILL.md.
Links to these hosts (documentation or services it may open):
claude.aiFrom URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Orchestrix Guide loads about 4k tokens when it runs. Until then it costs about 115 tokens; SKILL.md has 549 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from LeoYeAI/openclaw-master-skills at commit e5199b5, republished under its MIT licence (© LeoYeAI). 549 words, ~3,969 tokens.
.claude/skills/orchestrix-guide/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.本文档是 OpenClaw 操控 Orchestrix Agent 的唯一操作手册。 严格按照本文档的阶段和协议执行,不得跳步。
OpenClaw (自动化控制层,接收用户指令: Telegram / WhatsApp / Slack)
↓
tmux (终端复用层) ← OpenClaw 通过 tmux send-keys 发送命令
↓
Claude Code - cc (AI 编码助手,交互式 CLI,无 HTTP API)
↓ 通过 /o 激活 Agent
Orchestrix MCP Server → 返回 Agent 配置和工作流
↓
Claude Code 执行 Agent 工作流 → 输出文档/代码/HANDOFF关键约束:Claude Code (cc) 只接受终端标准输入。OpenClaw 唯一的操控方式是 tmux send-keys。
每次向 Agent 发送任务前,必须完整执行此三步。不得跳过任何一步。
WIN="{session}:{window}"
# Step 1: 清空上下文(防止角色残留和混乱)
tmux send-keys -t $WIN "/clear" Enter
sleep 2
# Step 2: 激活目标 Agent
tmux send-keys -t $WIN "/o {agent}" Enter
sleep 3
# Step 3: 发送任务指令
tmux send-keys -t $WIN "*{command}" Enter唯一例外:同一 Agent 的连续对话中追加指令(不切换 Agent),可省略 Step 1-2。
| 操作 | 等待时间 | 原因 |
|---|---|---|
cc 启动后 | 5s | 等待 Claude Code 初始化 |
/clear 后 | 2s | 等待上下文清空 |
/o {agent} 后 | 3s | 等待 Agent 加载和问候输出 |
*command 后 | 按任务检测 | 见下方完成检测策略 |
Agent 任务执行时间不确定,OpenClaw 必须准确判断任务是否完成。按优先级依次尝试:
tmux capture-pane -t {session}:{window} -p | grep -q "🎯 HANDOFF TO"适用场景:tmux 多窗口模式下所有任务。检测到即表示当前 Agent 已完成。
# 记录任务开始前时间戳
BEFORE=$(date +%s)
# ... 发送命令后轮询 ...
find ~/Codes/{project-name}/docs/ -newer /tmp/task-start-marker -type f | head -1适用任务:*create-doc *、*draft、*shard、*develop-story(有 git commit)等有文件产出的任务。
# Claude Code 执行完毕后显示 "Xxxed for <duration>"(如 "Worked for 33s")
tmux capture-pane -t {session}:{window} -p | grep -qE '[A-Z][a-z]+ed for'适用场景:所有任务。可作为 P1 的辅助确认。
PREV_HASH=""
STABLE_COUNT=0
while [ $STABLE_COUNT -lt 5 ]; do
HASH=$(tmux capture-pane -t {session}:{window} -p | md5)
if [ "$HASH" = "$PREV_HASH" ]; then
STABLE_COUNT=$((STABLE_COUNT + 1))
else
STABLE_COUNT=0
fi
PREV_HASH=$HASH
sleep 10
done连续 5 次以上 hash 不变才判定完成,防止 Agent 思考间隙被误判。
| 优先级 | 方法 | 可靠性 |
|---|---|---|
| P0 | HANDOFF 指令 | 最高(仅多窗口模式) |
| P1 | 预期输出文件 | 高 |
| P2 | 终端完成标志 [A-Z][a-z]+ed for | 高 |
| P3 | 终端内容 hash 稳定 | 中(兜底) |
建议组合:规划阶段用 P1 + P2 双重确认;开发阶段用 P0 + P2 为主。
一个 Orchestrix 项目的完整生命周期分为两大阶段。
┌─────────────────────────────────────────────────────────┐
│ Phase A: 规划阶段 │
│ (单窗口模式,逐个切换 Agent) │
│ │
│ 前提条件: 项目已通过 /create-project 创建 │
│ docs/project-brief.md 已存在(初版) │
│ │
│ Step 0: Analyst → *create-doc project-brief (可选深化) │
│ Step 1: PM → *create-doc prd │
│ Step 2: UX Expert → *create-doc front-end-spec (可选) │
│ Step 3: Architect → *create-doc fullstack-architecture │
│ Step 4: PO → *execute-checklist + *shard │
│ │
│ ✅ 规划完成标志: PO *shard 执行完毕 │
└─────────────────────┬───────────────────────────────────┘
↓
┌─────────────────────────────────────────────────────────┐
│ Phase B: 开发阶段 │
│ (多窗口模式,HANDOFF 自动协作) │
│ │
│ 启动: bash .orchestrix-core/scripts/start-orchestrix.sh │
│ │
│ SM *draft → Architect *review → Dev *develop-story │
│ → QA *review → SM *draft (下一个) → ... 循环 │
│ │
│ ✅ 开发完成标志: 所有 Story 通过 QA 审查 │
└─────────────────────────────────────────────────────────┘规划阶段在一个 tmux 窗口中完成,通过逐个切换 Agent 生成全套规划文档。
SESSION="orchestrix-planning"
PROJECT_DIR=~/Codes/{project-name}
# 创建 session 并启动 Claude Code
tmux new-session -d -s $SESSION -c $PROJECT_DIR
tmux send-keys -t $SESSION:0 "cc" Enter
sleep 5此步骤可选。
docs/project-brief.md已由/create-project生成初版(含问题陈述、目标用户、MVP 功能、技术栈),PM 可直接使用。 Analyst 的作用是在初版基础上深化(加市场调研、竞品分析),适合对项目背景需要更深入了解的场景。
WIN="$SESSION:0"
tmux send-keys -t $WIN "/o analyst" Enter
sleep 3
tmux send-keys -t $WIN "*create-doc project-brief" Enter
# 等待完成(P1: 检查 docs/project-brief.md 更新时间 + P2: 终端完成标志)预期产出:docs/project-brief.md(深化版)
这是规划阶段的必要起点。 PM 基于已有的
docs/project-brief.md(无论初版还是深化版)生成产品需求文档。
# 如果前面执行了 Step 0,需要先 /clear;如果直接从 Step 1 开始,这是第一条命令
tmux send-keys -t $WIN "/clear" Enter
sleep 2
tmux send-keys -t $WIN "/o pm" Enter
sleep 3
tmux send-keys -t $WIN "*create-doc prd" Enter
# 等待完成(P1: 检查 docs/prd/ 目录下新文件 + P2)预期产出:docs/prd/*.md(产品需求文档)
仅当项目有前端需求时执行。 纯后端/CLI 项目跳过此步。
tmux send-keys -t $WIN "/clear" Enter
sleep 2
tmux send-keys -t $WIN "/o ux-expert" Enter
sleep 3
tmux send-keys -t $WIN "*create-doc front-end-spec" Enter
# 等待完成(P1: 检查 docs/ 下前端规格文件 + P2)预期产出:docs/front-end-spec*.md
tmux send-keys -t $WIN "/clear" Enter
sleep 2
tmux send-keys -t $WIN "/o architect" Enter
sleep 3
tmux send-keys -t $WIN "*create-doc fullstack-architecture" Enter
# 等待完成(P1: 检查 docs/ 下架构文件 + P2)预期产出:docs/architecture*.md
这是规划阶段的最后一步。PO 需要连续执行两个命令。
# 4a: PO 验证文档一致性
tmux send-keys -t $WIN "/clear" Enter
sleep 2
tmux send-keys -t $WIN "/o po" Enter
sleep 3
tmux send-keys -t $WIN "*execute-checklist po-master-validation" Enter
# 等待完成(P2: 终端完成标志)
# 4b: PO 分片文档(同一 Agent,无需 /clear)
tmux send-keys -t $WIN "*shard" Enter
# 等待完成(P1: 检查 .orchestrix-core/context/ 或 docs/ 下分片文件 + P2)预期产出:
PO *shard 执行完毕后,规划阶段结束。销毁规划会话:
tmux kill-session -t $SESSION此时项目 docs/ 目录应包含:
project-brief.md — 深化后的项目简报prd/*.md — 产品需求文档front-end-spec*.md — 前端规格(如适用)architecture*.md — 架构文档✅ 规划完成。可以进入 Phase B 开发阶段。
#!/bin/bash
# OpenClaw 规划阶段自动化脚本
# 用法: bash planning.sh {project-name} [--with-analyst]
# --with-analyst 先让 Analyst 深化项目简报(可选)
PROJECT_NAME="$1"
WITH_ANALYST="$2"
PROJECT_DIR=~/Codes/$PROJECT_NAME
SESSION="orchestrix-planning"
WIN="$SESSION:0"
# 检查项目和 project-brief 是否存在
if [ ! -f "$PROJECT_DIR/docs/project-brief.md" ]; then
echo "ERROR: $PROJECT_DIR/docs/project-brief.md not found. Run /create-project first."
exit 1
fi
# 启动 tmux + cc
tmux new-session -d -s $SESSION -c $PROJECT_DIR
tmux send-keys -t $WIN "cc" Enter
sleep 5
# --- Step 0: Analyst (可选) ---
if [ "$WITH_ANALYST" = "--with-analyst" ]; then
tmux send-keys -t $WIN "/o analyst" Enter
sleep 3
tmux send-keys -t $WIN "*create-doc project-brief" Enter
# → OpenClaw 在此等待任务完成(P1 + P2 检测)
tmux send-keys -t $WIN "/clear" Enter && sleep 2
fi
# --- Step 1: PM ---
tmux send-keys -t $WIN "/o pm" Enter && sleep 3
tmux send-keys -t $WIN "*create-doc prd" Enter
# → 等待完成
# --- Step 2: UX Expert (可选,根据项目技术栈判断) ---
# if [ "$HAS_FRONTEND" = "true" ]; then
# tmux send-keys -t $WIN "/clear" Enter && sleep 2
# tmux send-keys -t $WIN "/o ux-expert" Enter && sleep 3
# tmux send-keys -t $WIN "*create-doc front-end-spec" Enter
# # → 等待完成
# fi
# --- Step 3: Architect ---
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o architect" Enter && sleep 3
tmux send-keys -t $WIN "*create-doc fullstack-architecture" Enter
# → 等待完成
# --- Step 4: PO 验证 + 分片 ---
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o po" Enter && sleep 3
tmux send-keys -t $WIN "*execute-checklist po-master-validation" Enter
# → 等待完成
tmux send-keys -t $WIN "*shard" Enter
# → 等待完成
# 规划完成
echo "✅ Planning phase complete for $PROJECT_NAME"
echo "Next: cd $PROJECT_DIR && bash .orchestrix-core/scripts/start-orchestrix.sh"
tmux kill-session -t $SESSION开发阶段通过 tmux 多窗口模式运行,4 个 Agent 通过 HANDOFF 自动协作,无需人工切换。
*shard 已执行)docs/ 目录下有完整的规划文档cd ~/Codes/{project-name}/
bash .orchestrix-core/scripts/start-orchestrix.sh脚本自动完成:
orchestrix-{repo-id}cc*draft 开始第一个 Story| 窗口 | Agent | 职责 |
|---|---|---|
0 | Architect | 技术审查、架构守护 |
1 | SM | Story 创建、流程编排 |
2 | Dev | 代码实现 |
3 | QA | 代码审查、质量验证 |
SM (窗口1) *draft → 创建 Story
↓ 🎯 HANDOFF TO architect: *review {story_id}
Architect (窗口0) → 技术审查
↓ 🎯 HANDOFF TO dev: *develop-story {story_id}
Dev (窗口2) → 编码实现
↓ 🎯 HANDOFF TO qa: *review {story_id}
QA (窗口3) → 代码审查
↓ 🎯 HANDOFF TO sm: *draft (下一个 Story)
SM (窗口1) → 创建下一个 Story
↓ ... 循环直到所有 Story 完成handoff-detector.sh 自动完成:
🎯 HANDOFF TO {agent}: *{command}/clear + 重新加载 Agent# 实时查看 HANDOFF 日志
tail -f /tmp/orchestrix-{repo-id}-handoff.log
# 查看各窗口当前输出
tmux capture-pane -t orchestrix-{repo-id}:1 -p | tail -10 # SM
tmux capture-pane -t orchestrix-{repo-id}:2 -p | tail -10 # Dev
# 查看 Story 完成情况
ls ~/Codes/{project-name}/docs/stories/
# 查看 git 提交记录
cd ~/Codes/{project-name}/ && git log --oneline -10| 情况 | OpenClaw 操作 |
|---|---|
| Agent 卡住不输出 HANDOFF | tmux send-keys -t {session}:{window} "/clear" Enter → sleep 2 → /o {agent} |
| HANDOFF 没被检测到 | tail -20 /tmp/orchestrix-{repo-id}-handoff.log 排查 |
| 需要暂停 | tmux send-keys -t {session}:{window} C-c |
| 需要恢复 | tmux send-keys -t {session}:1 "*draft --continue" Enter |
| 需要完全停止 | tmux kill-session -t orchestrix-{repo-id} |
当需要额外 Agent 并行工作时(如 UX Expert),可动态创建窗口:
SESSION="orchestrix-{repo-id}"
# 创建新窗口
tmux new-window -t $SESSION -c ~/Codes/{project-name}/
NEW_WIN=$(tmux list-windows -t $SESSION -F '#{window_index}' | tail -1)
# 启动 cc 并加载 Agent
tmux send-keys -t $SESSION:$NEW_WIN "cc" Enter
sleep 5
tmux send-keys -t $SESSION:$NEW_WIN "/o ux-expert" Enter
sleep 3
tmux send-keys -t $SESSION:$NEW_WIN '*create-doc front-end-spec' Enter
# 任务完成后销毁临时窗口
# tmux kill-window -t $SESSION:$NEW_WIN注意:动态窗口不参与 HANDOFF 自动协作,需 OpenClaw 自行管理生命周期。
跳过 Story/QA 关卡,适用于小型独立项目或快速原型。
# 单窗口即可
tmux send-keys -t $WIN "/o dev" Enter && sleep 3
tmux send-keys -t $WIN '*solo "实现用户登录功能,支持邮箱和手机号"' EnterDev 自动完成:创建 Story → 编码 → 测试 → 提交。
轻量修复(不需要 Story):
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o dev" Enter && sleep 3
tmux send-keys -t $WIN '*quick-fix "登录页面在 Safari 下白屏"' Enter正式修复(需要追踪):
# SM 创建 bugfix Story
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o sm" Enter && sleep 3
tmux send-keys -t $WIN '*draft-bugfix "用户并发下单时库存出现负数"' Enter
# → 等待完成,获取 story_id
# Dev 开发修复
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o dev" Enter && sleep 3
tmux send-keys -t $WIN "*develop-story {bugfix_story_id}" Enter
# QA 验证
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o qa" Enter && sleep 3
tmux send-keys -t $WIN "*review {bugfix_story_id}" Enter
tmux send-keys -t $WIN "*finalize-commit {bugfix_story_id}" Enter当一个 Epic 下所有 Story 都通过 QA 审查后执行。
# 单窗口模式
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o qa" Enter && sleep 3
tmux send-keys -t $WIN "*smoke-test {epic_id}" Enter
# 多窗口模式(直接发到 QA 窗口)
tmux send-keys -t orchestrix-{repo-id}:3 "*smoke-test {epic_id}" Enter如果冒烟测试发现问题 → 走 6.2 正式修复流程 → 再次冒烟测试。
MVP 完成后,基于反馈启动新一轮迭代。
# Step 1: PM 生成 next-steps
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o pm" Enter && sleep 3
tmux send-keys -t $WIN "*start-iteration" Enter
# → 等待完成,产出 docs/prd/*next-steps.md
# Step 2: OpenClaw 读取 next-steps.md,解析 🎯 HANDOFF TO 块
# Step 3: 按文件中的 HANDOFF 顺序依次执行
# 通常顺序:ux-expert → architect → sm
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o ux-expert" Enter && sleep 3
tmux send-keys -t $WIN "{HANDOFF TO ux-expert 部分内容}" Enter
# → 等待完成
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o architect" Enter && sleep 3
tmux send-keys -t $WIN "{HANDOFF TO architect 部分内容}" Enter
# → 等待完成
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o sm" Enter && sleep 3
tmux send-keys -t $WIN "{HANDOFF TO sm 部分内容}" Enter
# → 等待完成
# Step 4: 启动多窗口自动化开发(同 Phase B)
cd ~/Codes/{project-name}/
bash .orchestrix-core/scripts/start-orchestrix.sh# 所有变更先经 PO 路由
tmux send-keys -t $WIN "/clear" Enter && sleep 2
tmux send-keys -t $WIN "/o po" Enter && sleep 3
tmux send-keys -t $WIN "*route-change" EnterPO 根据变更类型自动路由:
*revise-prd)*resolve-change)| 变更规模 | 推荐方式 |
|---|---|
| < 1h 快速修复 | /o dev → *quick-fix |
| < 4h 单功能 | /o sm → *draft |
| 4h-2d 小型增强 | /o sm → *draft(brownfield epic) |
| > 2d 大型增强 | 走完整 Phase A → Phase B 流程 |
对不熟悉的项目先摸底:/o architect → *document-project
| Agent | ID | 核心命令 | 产出 |
|---|---|---|---|
| Analyst | analyst | *create-doc project-brief | docs/project-brief.md |
| PM | pm | *create-doc prd, *revise-prd, *start-iteration | docs/prd/*.md |
| UX Expert | ux-expert | *create-doc front-end-spec | docs/front-end-spec*.md |
| Architect | architect | *create-doc fullstack-architecture, *document-project | docs/architecture*.md |
| PO | po | *execute-checklist po-master-validation, *shard | 验证报告 + 分片文件 |
| Agent | ID | 核心命令 | 产出 |
|---|---|---|---|
| SM | sm | *draft, *draft-bugfix {bug} | docs/stories/*.md |
| Architect | architect | *review {story_id} | 技术审查意见 |
| Dev | dev | *develop-story {story_id}, *solo "{desc}", *quick-fix "{desc}" | 代码 + git commit |
| QA | qa | *review {story_id}, *finalize-commit {story_id}, *smoke-test {epic_id} | 审查报告 + 最终提交 |
| Agent | ID | 核心命令 |
|---|---|---|
| PO | po | *route-change |
| Orchestrator | orchestrix-orchestrator | *status, *workflow-guidance |
| 依赖 | 用途 | 安装 |
|---|---|---|
claude (Claude Code) | AI 编码环境 | https://claude.ai/download |
tmux | 多窗口终端复用(必须) | brew install tmux |
git | 版本控制 | 系统自带 |
jq | JSON 处理(可选) | brew install jq |
别名配置(start-orchestrix.sh 依赖):
alias cc='claude --dangerously-skip-permissions'© LeoYeAI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 1 other file in skills/orchestrix-guide of LeoYeAI/openclaw-master-skills.
Open the folder on GitHubat commit e5199b5
Orchestrix Guide next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Orchestrix Guide this skillLeoYeAI/openclaw-master-skills | 2.2k | — | ~4k | Automated safety check: Pass | MIT | |
| Agent of Empires Session Manageragent-of-empires/agent-of-empires | 3.3k | — | ~2.1k | Automated safety check: Pass | MIT | |
| Agent Manager Fleet TUIYoanWai/agent-manager | 578 | — | ~1k | Automated safety check: Pass | Apache-2.0 | |
| Agtx Task Sweepfynnfluegge/agtx | 1.7k | — | ~1.7k | Automated safety check: Pass | Apache-2.0 | |
| Error Debugging Multi Agent Reviewaiskillstore/marketplace | 430 | 7 repos | ~1.8k | Automated safety check: Pass | None | |
| Agent of Empires Session Manageragent-of-empires/agent-of-empires | 3.3k | — | ~2.5k | Automated safety check: Pass | MIT |
agent-of-empires/agent-of-empires
Starts, monitors and organizes coding agent sessions that run in tmux through the aoe command, including groups, profiles and worktree-based parallel work.
YoanWai/agent-manager
Runs several coding-agent CLIs as real tmux sessions in one terminal UI, color-coded by whether each is working, waiting, idle or blocked.
fynnfluegge/agtx
Breaks a conversation's results into feature-level tasks and pushes them to the agtx kanban board, where each task gets its own worktree and agent session.
aiskillstore/marketplace
A skill your agent uses when working with error debugging multi agent review
agent-of-empires/agent-of-empires
Launches, monitors and organizes AI coding agent sessions such as Claude Code or Codex inside tmux, tracking status, capturing output and managing git worktrees for parallel branches.
win4r/ClawTeam-OpenClaw
Multi-agent swarm orchestration. An agent skill from win4r/ClawTeam-OpenClaw.
LeoYeAI/openclaw-master-skills
Manages pipelines on a DevOps quality and efficiency platform through its OpenAPI: list workspaces and templates, create, update, run and cancel pipelines, and read run records.
LeoYeAI/openclaw-master-skills
Patches OpenClaw's Feishu extension so an edited document triggers an isolated agent session that reads the doc and replies inline, turning it into a live chat space.
LeoYeAI/openclaw-master-skills
Multi-context memory management system for OpenClaw agents with group-isolated storage, global shared memory, workspace organization, and group-specific skills isolation.
LeoYeAI/openclaw-master-skills
Runs a brand's AI-search visibility work end to end: diagnosing how AI platforms represent it, repositioning it, producing AI-optimized content and monitoring ongoing mentions.
LeoYeAI/openclaw-master-skills
Installs and authenticates the gws CLI, then automates Gmail, Drive, Sheets, Calendar, Docs, Chat and Tasks with ready-made recipes, persona bundles and security audits.
LeoYeAI/openclaw-master-skills
Runs four advisor roles, a fitness coach, nutritionist, data analyst and TCM practitioner, to build a health profile and track workouts, diet and wellness over time.
Works with
Orchestrix multi-agent workflow guide for OpenClaw. An agent skill from LeoYeAI/openclaw-master-skills. Orchestrix Guide is an agent skill from LeoYeAI/openclaw-master-skills. Orchestrix multi-agent workflow guide for OpenClaw.
Orchestrix Guide fits situations like: tasks that involve PRD writing; tasks that involve Multi-agent orchestration; tasks that involve Debugging.
Run `npx skills add LeoYeAI/openclaw-master-skills --skill orchestrix-guide -a claude-code`. Or copy the skill folder (skills/orchestrix-guide in LeoYeAI/openclaw-master-skills) into .claude/skills/orchestrix-guide in your project. Claude Code loads it when a task matches its description.
Run `npx skills add LeoYeAI/openclaw-master-skills --skill orchestrix-guide -a codex`. Or copy the skill folder (skills/orchestrix-guide in LeoYeAI/openclaw-master-skills) into .agents/skills/orchestrix-guide in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add LeoYeAI/openclaw-master-skills --skill orchestrix-guide -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/orchestrix-guide, .gemini/skills/orchestrix-guide, .github/skills/orchestrix-guide and .opencode/skills/orchestrix-guide in your project.
Going by SKILL.md and its folder, Orchestrix Guide needs the command-line tools its instructions call (bash, claude, brew and git).
SKILL.md names 1 domain. As links in the text: claude.ai. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Orchestrix Guide is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.
About 4k tokens (SKILL.md is roughly 16k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Orchestrix Guide: Agent of Empires Session Manager (agent-of-empires/agent-of-empires, 3.3k stars), Agent Manager Fleet TUI (YoanWai/agent-manager, 578 stars), Agtx Task Sweep (fynnfluegge/agtx, 1.7k stars) and Error Debugging Multi Agent Review (aiskillstore/marketplace, 430 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
LeoYeAI (a GitHub user) maintains it in LeoYeAI/openclaw-master-skills, which has 2,160 GitHub stars. The repository holds 1,235 skills in this directory. The repository was last updated on July 20, 2026.
Source: LeoYeAI/openclaw-master-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.