multi-agent-orchestrator
Skill1 starsChanged 3 months agoApache-2.0View on GitHub
在线多智能体对话总指挥。DeepSeek 主脑通过 CDP 调度夸克浏览器里的千问(p.quark.cn/pcquark-chat/sidebar)与豆包(doubao.com/chat)等在线智能体,并行回答同一问题、收集数据,再提取结构化结果并用本地数据验证。支持文字/图片附件;视频走抽帧、音频暂不支持(转文字替代)。当用户要"让千问/豆包一起回答"、需要在线智能体收集资料/多视角答案、或要对比多家AI观点时使用。核心纪律:在线智能体只当"候选假设生成器",最终结论必须经本地数据/回测裁决。
10 other projects use a version of this.
Adopted by
Most recently taken up by
- hongphuc5497/skillstheir own version, first seen Nov 1, 2026
- Phat-Po/multi-agent-orchestratortheir own version, first seen Oct 15, 2026
- saintgo7/claude-skillstheir own version, first seen Oct 11, 2026
- iratus137-hue/multi-agent-orchestratortheir own version, first seen Jul 23, 2026
- tekliner/improvado-agentic-frameworks-and-skillstheir own version, first seen Jun 17, 2026
- lazze2026/multi-agent-orchestrator-skilltheir own version, first seen Jun 10, 2026
Other versions10 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 10
- egorfedorov/Slot-Casino-Game-Developer-Skills-for-Stake-Engine★ 68same name
- tekliner/improvado-agentic-frameworks-and-skills★ 17same name
- tellmefrankie/ai-investment-skills★ 14same name
- hongphuc5497/skills★ 2same name
- hyw007726/multi-agent-orchestrator-skill★ 2same name
- SKY-lv/multi-agent-orchestrator★ 2same name
- lazze2026/multi-agent-orchestrator-skill★ 1same name
- iratus137-hue/multi-agent-orchestrator★ 0same name
- Phat-Po/multi-agent-orchestrator★ 0same name
- saintgo7/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository51 other files it gives its agents.
- 3d-animation-short-generatorSkill
- brand-promo-video-generatorSkill
- colleague-cameronSkill
- colleague-muskSkill
- colleague-supermateSkill
- comfyui-zimage-9bSkill
- context-handoffSkill
- co-op-game-intro-generatorSkill
- deepseek-eyesSkill
- dot-skillSkill
- doubao-creatorSkill
- dsh-deepseek-vision-bridgeSkill
- dsh-im-wecomSkill
- cordis-plugin-developmentSkill
- dsh-archive-agent-notesSkill
- dsh-code-reviewSkill
- dsh-doc-site-syncSkill
- dsh-doc-standardsSkill
- dsh-find-simplificationsSkill
- dsh-merging-stacked-prsSkill
- dsh-pre-push-checksSkill
- dsh-prose-standardSkill
- dsh-translate-docsSkill
- dsh-trim-cot-leakageSkill
- editing-cordis-compositionsSkill
- record-browser-gifSkill
- dsh-upgrade-runbookSkill
- h3-prompt-writingSkill
- handdrawn-live-video-generatorSkill
- hermes-memorySkill
- image-analysisSkill
- lyric-mv-storyboardSkill
- lyric-to-storyboard-face-lockSkill
- lyric-to-storyboard-lrcSkill
- lyric-to-storyboardSkill
- minimalist-product-ad-generatorSkill
- h3-video-producerSkill
- modelscope-apiSkill
- music-video-subtitle-generatorSkill
- paper-collage-explainer-generatorSkill
- papercraft-stop-motion-explainerSkill
- quark-driveSkill
- quark-qwen-visionSkill
- rh-workflow-9bSkill
- rh-workflowSkill
- supermate-code-reviewSkill
- Supermate-harness-launcherSkill
- supermate-scriptSkill
- colleague-supermateSkill
- video-production-studioSkill
- visual-file-organizerSkill
The file
Files
在线多智能体对话总指挥(multi-agent-orchestrator)
模型矩阵:DeepSeek(主脑/裁决) + 千问(视觉/生图/知识) + 豆包(字节系/视频/检索) + 可扩展其他在线智能体。 铁律:智能体的回答是"假设",不是"结论"——必须落盘、提取、再用本地数据/脚本验证后才算数。
何时使用
- 用户想"让千问和豆包同时回答同一个问题"、对比多家 AI 观点
- 需要在线智能体联网收集资料、生成候选(如选股名单、资料综述、多视角分析)
- 问题带附件:文字/图片/视频/音频
依赖(已就绪)
| 组件 | 位置 |
| CDP 客户端 | quark-qwen-vision/scripts/cdp.js、doubao-creator/scripts/cdp.js(通用) |
| 千问提问/读回 | quark-qwen-vision/scripts/ask-qwen.js(发文字)、check-qwen.js(复查)、read-qwen.js(轮询)、qwen-vision.js(图片) |
| 豆包提问/看图 | doubao-creator/scripts/ask-doubao.js(发文字)、doubao-vision.js(图片) |
| 浏览器 | 夸克以 --remote-debugging-port=9222 运行;千问侧栏与豆包页面已打开 |
前置确认:http://127.0.0.1:9222/json/list 中能找到 pcquark-chat 与 doubao.com 页面。
总指挥工作流
1. 拆解任务 → 明确要给各智能体的"统一问题"(写入 logs\qwen_question.txt)
2. 并行派发:每个智能体一个后台 job 跑 node ask-*.js(同一问题文件)
3. 结果落盘:各智能体回复写入 logs\agent_<名>_<时间>.txt(原始全文不拉进对话)
4. 提取数据:只从回复中解析 表格/名单/要点(结构化部分),丢弃 UI 噪音
5. 本地验证:候选/结论用本地数据或脚本复核(例:validate_agents_pool.py 验证股票池)
6. 汇总交付:对比表 + 验证结果 + 结论;注明各智能体局限性并行派发示例
# 问题写入文件后,两个智能体各起一个后台任务同时回答:
node "…\quark-qwen-vision\scripts\ask-qwen.js" "E:\…\logs\qwen_question.txt" # 千问
node "…\doubao-creator\scripts\ask-doubao.js" "E:\…\logs\qwen_question.txt" # 豆包附件适配矩阵
| 附件 | 千问侧栏 | 豆包 | 说明 |
| 文字 | ask-qwen.js <文件> | ask-doubao.js <文件> | 问题写入 txt,规避命令行引号问题 |
| 图片 | qwen-vision.js <图> <问题> | doubao-vision.js <图> <问题> | 千问走 paste 事件;豆包走 DOM.setFileInputFiles |
| 视频 | 不支持直接粘贴 | 同上 | ffmpeg 抽帧 → 拼图/选关键帧 → 当图片喂 |
| 音频 | 不支持 | 不支持 | 先 ffmpeg 转文字,再走文字通道(标注局限) |
读取回复的坑(实战踩坑记录)
- 锚点截取:用"问题前 30-40 字"做锚点,只打印问题之后的回复;避免把历史对话拉进上下文
- 轮询稳定:innerText 连续 4-5 轮无变化才算回复完成;长任务(如选股)可能 3-6 分钟,耐心轮询(read-qwen.js 支持多页面)
- UI 噪音截断:豆包回复尾部截到"豆包 快速"等 UI 标记;千问侧栏 innerText 可能混入当前网页标题/上下文——提取时只信表格与结构化部分
- 千问侧栏"深度思考"可能超时:>6 分钟无结果不要傻等,改用
check-qwen.js复查或放弃该路 - 多页面歧义:pcquark-chat 可能有多个实例,用"问题锚点"定位真正收到提问的那个
数据提取约定(省 token 关键)
- 智能体长回复写文件,对话里只放结构化摘要(表格、名单、结论)
- 要求智能体输出"表格/编号清单",方便正则/规则提取
- 提取后的候选必须过本地验证脚本,输出只留"验证后"的结果
本地验证(裁决层)
- 股票类候选:
scripts/validate_agents_pool.py(是否成分股 + 箱体/趋势/波动达标)或box_scan_3y.py - 数据类结论:对照本地 data/ 与脚本结果
- 预测类:一律概率化 + trade-journal 复盘
实战复盘(2026-08-29 网格选股)
- 千问 10 只 + 豆包 14 只,两份名单完全无交集 → 在线智能体只适合粗筛
- 千问市值数据有硬伤(华能国际/大秦铁路实为千亿级,标注 ~180 亿)
- 本地三重验证:24 候选 → 8 只在成分股 → 2 只达标(601139 深圳燃气 / 600998 九州通)
- 教训固化:多智能体 = 快速生成假设 + 交叉视角;裁决永远用本地数据
关联技能
quark-qwen-vision(千问通道)·doubao-creator(豆包通道)·quant-data(本地数据)·astock-*(分析裁决)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

