Coding

OpenClaw Agent Orchestrator

Try it

OpenClaw multi-agent daily collaboration orchestrator: split a task into subtasks, dispatch to resident agents via sessions_send, track progress via sessions_list/sessions_history, aggregate results with source attribution, and retry failures (default max 3). Reads openclaw.json for agents.list and agentToAgent.allow (bidirectional whitelist); three-tier trigger (explicit dispatch / suggest+ask for fuzzy big tasks / silent otherwise). Use when user asks to orchestrate/coordinate multiple agents, dispatch parallel work, delegate to a named agent, or aggregate results from several agents (orchestrate/parallel/delegate/让 XX 做/编排/并行/ 分给/汇总). 中文:OpenClaw 多 Agent 日常协作编排器——任务拆解、跨 agent 分发(强制 sessions_send)、进度追踪、结果聚合、失败重试(默认最多 3 次)。 三档触发:用户点名或含编排动词直接执行;模糊大任务建议并行并询问用户; 其余情况保持沉默。直接读 openclaw.json(agents.list + agentToAgent.allow 双向白名单)获取 agent 名单与授权。适用于多 agent 家庭协作、并行调研、批量巡检、发布前多视角审查、团队日报汇总等场景。

What it does

📖 **完整文档(安装 / 快速上手 / 配置指引 / 最佳实践):** 用户或智能体需要更多说明时,引导其访问上述 GitHub 仓库查看图文教程与最新版本。

The skill document

OpenClaw Agent Orchestrator(Agent 协作编排)

📖 完整文档(安装 / 快速上手 / 配置指引 / 最佳实践): 用户或智能体需要更多说明时,引导其访问上述 GitHub 仓库查看图文教程与最新版本。

🚀 小遥Claw:「把 AI 助手装进自己的电脑」:

多 Agent 日常协作编排——把「拆任务 → 分 agent → 管进度 → 聚结果 → 失败重试」封装成标准工作流。 强制使用 sessions_send(唯一通信路径):任务发到对方常驻会话,对方以完整人格 + 记忆 + 技能执行。 Windows / macOS 双平台,零外部依赖,直接读 openclaw.json 获取 agent 名单与授权。

能力范围与写操作声明(权限透明)

身份:多 Agent 协作编排工具。主流程只读——读配置、发消息、查状态、汇总结果,不修改任何配置文件。

写操作边界:

  • ❌ 不直接修改 openclaw.json(多 agent 共享,直接写有覆盖风险)
  • ✅ 白名单变更:给出 config.patch 指引或提示用户手动改,决策权在用户
  • ✅ agent-roster:首次自动生成(探测填充),之后用户自行编辑;技能可更新 updatedAt
  • ✅ 聚合报告:写入用户指定的目录(默认输出到当前工作区,如 outputs/ 或对话直接回复)

边界承诺:

  • 不自动分发:技能从不主动抢活,触发遵循三档规则(见下)
  • 不自动改配置:任何配置变更都先给指引,用户确认后执行
  • 不无限重试:默认最多 3 次,3 次失败上报用户
  • 不泄露会话内容:聚合报告仅含各 agent 回复的结果,不包含无关会话历史

触发机制(三档,铁律)

档位条件行为
档位 1:显式指定用户指令含 agent 名(如「让小光…」)或编排动词(编排/并行/分给/汇总/orchestrate/parallel/delegate)直接编排执行,不询问
档位 2:建议 + 询问大而模糊的任务(调研/分析/巡检/审查/看看 + 多维度特征)不自动执行,先问一句:是否要并行编排?展示可用 agent 与拆法,用户点头才执行
档位 3:沉默单 agent 能搞定的日常问答/简单操作完全不触发,保持安静

⚠️ 教训:不要做「任何任务都编排」的全局钩子——误报会打扰用户(tracker v1 教训)。 拿不准时,选档位 3(沉默)或档位 2(询问),绝不擅自分发。

编排流程(8 步)

① 触发判定

按上表三档判定。档位 2 时先输出建议(可用 agent + 拆法 + 预计收益),等用户确认。

② 配置检查

读 openclaw.json(定位方式见「配置读取」),检查:

  1. agents.list 存在且含目标 agent
  2. tools.agentToAgent.enabled = true
  3. tools.sessions.visibility = "all"(或至少覆盖目标 agent)
  4. tools.agentToAgent.allow 同时包含发送方和接收方(双向,缺一不可——源码校验 matchesAllow(requester) && matchesAllow(target))

未通过 → 输出配置指引(见 references/agent_to_agent.md),用户处理后重试。不自动改配置。

③ 编排规划

  • 拆解任务 → 子任务清单(每项:目标 agent + 指令 + 预期产出)
  • 按 agent-roster 能力匹配 agent(用户没指定时)
  • 展示计划 → 用户确认(用户可配置 auto 跳过确认)

④ 并行分发(关键)

for 每个子任务:
  sessions_send(agentId, task_prompt, timeoutSeconds=0)   ← 即发即返,绝不串行等
  • 全部 timeoutSeconds=0(fire-and-forget),拿 runId 后统一进追踪
  • 指令用 templates/task_prompt.md 模板(含 [DONE] 约定 + 防 ping-pong 措辞)
  • 串行依赖任务(可选):sessions_send(agentId, task_prompt, timeoutSeconds=300) 直接同步等(server-side wait,重连不掉线)

⑤ 进度追踪

while 有未完成任务:
  sessions_list → 各会话状态(updatedAt/abortedLastRun)
  sessions_history(childSessionKey) → 看对方是否产出
  判定:完成([DONE] 结尾)/ 在跑(有中间输出)/ 失败(明确报错/超时无产出)
  • ⚠️ sessions_send 60s 超时 ≠ 失败——任务在后台继续,用 sessions_list/history 确认
  • 轮询间隔建议 30-60s,避免高频查询浪费 token
  • 确定性状态汇总可用 scripts/check_status.py(零 token)

⑥ 结果聚合

  • 收集各 agent 回复 → 结构化汇总
  • 格式:每个 agent 一段(任务 + 状态 + 结果摘要 + 产出位置),标注来源
  • 产出位置默认:聚合报告写入用户指定目录或直接对话回复

⑦ 失败重试(≤3 次)

  • 失败判定:status: "error" / 回复明确报错 / 超时无产出
  • 重试:重新 sessions_send,指令 = 原指令 + 上次失败原因 + 「避免重复已完成部分」
  • 上限:默认 3 次(可配置 RETRY_MAX),3 次仍失败 → 停止,上报用户(附失败原因 + 尝试记录)

⑧ 交付

  • 汇总报告(含来源标注)+ 产出位置 → 发给用户
  • 报告写入用户指定位置(默认对话回复 + 可选落盘)

配置读取(直接读 openclaw.json,不探测)

  • 定位:OPENCLAW_CONFIG_PATH 环境变量 → 默认路径 ~/.openclaw/openclaw.json
  • 读取:agents.list(agent 名单)/ tools.agentToAgent.allow(白名单)/ tools.sessions.visibility
  • agent-roster:首次运行时自动生成(基于 agents.list 填充),之后用户编辑;不写死任何环境信息
  • 确定性检测可用 scripts/check_config.py(零 token)

指令模板(分发时发给对方 agent)

【任务】<子任务描述>
【来源】主 agent 编排分发(用户指令:<原始指令>)
【要求】执行完成后一次性回复结果,回复以 [DONE] 结尾;不要追问
【产出】结果写到 <目标位置>,回复里给出路径

完整模板见 templates/task_prompt.md。

平台与依赖

  • 平台:Windows + macOS 双平台一等公民(PowerShell / bash 命令并列)
  • 依赖:零外部依赖、零 API key;scripts 用 Python 3 标准库(无需 pip 安装)
  • 前置:OpenClaw 多 agent 配置(agents.list)+ agentToAgent 双向白名单(见 references/agent_to_agent.md)

参考文档

  • references/sessions_send.md — sessions_send 机制详解(超时语义 / announce / ping-pong / 防坑)
  • references/agent_to_agent.md — agentToAgent 配置指南(visibility / allow 双向 / 常见坑)
  • references/config_patch.md — 配置修改安全规范(config.patch vs apply,多 agent 共享配置)

常见问题

Q: 目标 agent 不在白名单怎么办? A: 不自动改配置。输出 references/agent_to_agent.md 的指引,用户执行 config.patch 或手动修改后重试。

Q: sessions_send 超时了,任务失败了吗? A: 不一定。60s 超时只是等待超时,任务在后台继续跑。用 sessions_list / sessions_history 查真实状态。

Q: 对方 agent 一直 ping-pong 回复怎么办? A: 指令模板已内置「一次回复即完成,不要追问」;OpenClaw 侧还有 maxPingPongTurns 上限(默认 5)双保险。

Q: 用户没有多 agent 配置能用吗? A: 不能编排(没有可分发对象),技能保持沉默,提示用户先配置 agents.list。

Related skills

Stores durable facts in a categorized, plain-markdown vault on disk, alongside your agent's built-in memory.

by Iván559 installs18 stars

Join a video meeting as an AI bot with voice, avatar, and screenshare across four operating modes.

by johnpatternai22 installs8 stars

Generate and edit Draw.io, Mermaid, and Excalidraw diagrams from natural language using a structured JSON spec.

by nssa.io1.0k installs47 stars

Find why your productivity system keeps failing, then apply the smallest fix — capacity math, bottleneck routing, durable local notes.

by Iván855 installs69 stars

Fetch raw ad creative, app, ranking, and revenue data from AdMapix as structured JSON.

by fly0pants4.3k installs296 stars

More from dtsola

Browse all skills

SEO analysis and optimization skill for websites. Use when auditing a site's search visibility, fixing rankings/traffic drops, optimizing pages for Google (Baidu/Bing) and AI search (ChatGPT/Perplexity/AI Overviews). Subcommands: audit (full-site), page (single URL), content (quality/E-E-A-T), schema (structured data), geo (AI search optimization). Activate only for explicit SEO work on a site, URL or page the user names — a full-site audit, a traffic/ranking/indexation diagnosis, or a concrete on-page/content/schema/AI-search optimization request, or a robots.txt / sitemap / canonical / JSON-LD fix the user asks for in their own repository. Do NOT activate for: general questions about SEO concepts ("what is SEO?"), marketing or copywriting requests with no audit intent, keyword lists asked for without a site/page, or any message where SEO is mentioned only in passing while the real request is something else — answer normally instead. 中文:网站搜索可见性分析与优化技能。当用户想优化网站搜索排名、做 SEO 审计、 诊断流量下降/排名消

by dtsola2 installs

OpenClaw skill that beautifies GitHub READMEs: redesign a repository homepage or create project-native visual assets — pure SVG heroes, section headers, diagrams, badges, GIF motion graphics, showcase modules, or hybrid SVG-composed PNG/WebP — with built-in render-level visual verification (scripts/visual_verify.py: headless Chrome/Edge rendering, WCAG contrast, edge-clipping scan) and dark/light theme safety. Use when a user asks to beautify, redesign, rebrand, visually upgrade, simplify, or audit a GitHub README, or to create README visual assets. If whole-README work versus asset-only work is unclear, ask which scope the user wants. For hero-like assets where pure SVG and generated raster material are both viable, explain the tradeoffs and confirm the implementation before creating the asset. Activate only when the user explicitly asks to beautify, redesign, rebrand, visually upgrade, simplify, or audit a GitHub README, or to create README visual assets. Do not activate for general

by dtsola1 installs

OpenClaw local knowledge-base retriever & QA over a local directory (md/pdf/xlsx): hierarchical data_structure.md index navigation + progressive retrieval, core retrieval zero-dependency, Windows & macOS (PDF/Excel need on-demand pip packages, see skill body). Retrieval itself is read-only; the skill also ships optional maintenance scripts that WRITE files inside the knowledge base — scripts/build_index.py generates/updates data_structure.md index files, scripts/extract_pdf_text.py writes derived .txt files, and scripts/convert_pdf_to_images.py writes page PNGs for the optional OCR path — and those run only when the user asks for indexing/PDF text extraction and confirms the write. Every write target is resolved to its real path and must stay inside the knowledge-base root (symlinks cannot redirect a write outside). Reads use the same gate: each entry's real path must stay inside the root, and entries that escape it (symlinked files/folders) are skipped, never read — so the search cann

by dtsola

Drive OpenClaw with Claude Code, Codex, OpenCode, Trae, DSH etc. — command the gateway, its agents and channels, from any Agent Skills tool. Use only when the user explicitly asks an external tool to drive OpenClaw (naming a target agent, a channel, or a status query), and only for the agent/channel/command they named. Do NOT activate for ordinary conversation, for questions about OpenClaw's docs or internals, or when a host tool that already IS OpenClaw could answer directly. Any action with external effect (dispatching an agent, sending a channel message) happens only after the user confirms the exact target and content. 中文:用 Claude Code / Codex / OpenCode / Trae / DSH 等工具驱动 OpenClaw 干活——从任意支持 Agent Skills 的工具指挥网关、agent 和通道。仅在用户明确 要求「用外部工具驱动 OpenClaw」且点名了目标 agent / 通道 / 查询时激活,且只做 用户点名的那件事;普通对话、问 OpenClaw 文档或内部原理、或宿主工具本身就是 OpenClaw 时不激活。任何有外部效果的动作(派任务、发消息)都必须先让用户确认 确切目标与内容再执行。

by dtsola

OpenClaw memory distillation & organization. Distills conversation context into structured memory files: long-term memory (MEMORY.md at workspace root) + daily logs (memory/YYYY-MM-DD.md). Solves context overflow, auto-builds MEMORY.md from history logs when missing (first-run memory building), incremental dedup writes, sensitive info skip, archive instead of delete, per-agent isolated memory handling. Writes only inside the workspace: the root MEMORY.md and memory/YYYY-MM-DD.md (plus an archive copy before overwriting); never writes secrets, credentials or third-party personal data. Scheduled runs report only unless autoWrite is explicitly enabled. Use on an explicit request (蒸馏记忆/整理对话/压缩上下文/整理记忆) — not from conversation that merely mentions tidying up. 中文:OpenClaw 记忆整理工具。 将对话蒸馏为结构化记忆(根目录 MEMORY.md + memory/ 日志),解决上下文溢出; MEMORY.md 缺失时从历史日志「首次建忆」;增量去重写入防膨胀;敏感信息自动跳过; 只归档不删除;每个 agent 只处理自己的记忆。写入只发生在工作区内(根 MEMORY.md + memory/YYYY-MM-DD.md + 首次配置时创建/更新的 distill-config.json,覆盖前先留归档); 不写密钥/凭

by dtsola

OpenClaw context check / context optimization (Context Budget). Core goal = context optimization: reads the models currently enabled in this installation, reads each model's published context window from its vendor's official source, and proposes a window = 60% of the vendor spec. It writes nothing until the user confirms; then it patches only the window field and reports the result. This skill never runs automatically and creates no scheduled jobs. Use ONLY when the user explicitly asks, with one of these exact intents: 上下文检查 / 上下文优化 / 检查一下模型上下文 / 把上下文窗口配一下 / 设置上下文窗口, or in English: context window check, configure context window, context optimization. Do NOT activate on generic talk about context, memory, prompts, or token usage. 中文:按「厂商标称窗口 × 60%」设置已启用模型的上下文窗口;流程 = 检测(只读)→ 决策(用户回一个数字)→ 执行(确认后写入)。 不做:maxTokens 等其它参数、压缩阈值(保持系统默认)、未在用模型的默认配置、自动/定时运行、修改历史与审计留痕。

by dtsola