Back to Skills

coordinate-swarm

pjt222
Updated 6 days ago
12 views
17
2
17
View on GitHub
Metaautomationdesign

About

The coordinate-swarm skill provides patterns like stigmergy and quorum sensing to design distributed systems that self-coordinate without a central controller. It helps developers build resilient, event-driven architectures by focusing on local agent rules and shared state communication. Use it to eliminate coordination bottlenecks and replace fragile orchestration with emergent, decentralized organization.

Quick Install

Claude Code

Recommended
Primary
npx skills add pjt222/agent-almanac -a claude-code
Plugin CommandAlternative
/plugin add https://github.com/pjt222/agent-almanac
Git CloneAlternative
git clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/coordinate-swarm

Copy and paste this command in Claude Code to install this skill

Documentation

調群

以 stigmergy(改境間通)、局互規、quorum 感,立分代間調——無中控而生協同群行。

  • 分系設、無單點為調瓶
  • 團工須自調而無常督
  • 事驅架、以共態通非直訊
  • 三代行而三十代潰之法
  • 新群域之調模引(見 forage-resourcesbuild-consensus
  • 代脆中組以韌生調

  • :須調代(工、服、員)之述
  • :群目或所求生行
  • :現調機與敗模
  • :代數(影模選——小群與大巢)
  • :延忍(實時或終至調)
  • :境限(共態可用、通頻寬)

一:別調問類

分調挑以選合模。

  1. 繪現態:代為誰、獨行為何、調於何處潰
  2. 分問:
    • Foraging — 代搜與用分資(見 forage-resources
    • Consensus — 代須同意一集決(見 build-consensus
    • Construction — 代漸築或守共構
    • Defense — 代群察與應威(見 defend-colony
    • Division of labor — 代須自組為專角
  3. 識現調之敗模:
    • 單點敗(中控)
    • 通瓶(直訊過多)
    • 協失(代無饋而漂)
    • 剛(不能適變)

得: 問類與所治具敗模之明分。此定施何群模。

敗: 問不合單類→可為合。分為子問、各以合模治。代過雜→考層調:同質簇以簇間 stigmergy 調。

二:設 stigmergic 信號

設代間接通道以相影行。

  1. 定共境(庫、訊隊、檔系、物空、共板)
  2. 設代沉於境之信號:
    • Trail signals:積於成徑之標(如蟻 pheromones)
    • Threshold signals:過閾觸行變之計
    • Inhibition signals:斥代離竭區之標
  3. 定信號性:
    • Decay rate:信號速衰(防陳態主)
    • Reinforcement:成果強信號
    • Visibility radius:信號傳遠
  4. 映信號至代行:
    • 代測信號 X 過閾 T→行動 A
    • 代成動 A→沉信號 Y
    • 無信號→代默探行
Signal Design Template:
┌──────────────┬───────────────────┬──────────────┬────────────────────┐
│ Signal Name  │ Deposited When    │ Decay Rate   │ Agent Response     │
├──────────────┼───────────────────┼──────────────┼────────────────────┤
│ success-trail│ Task completed OK │ 50% per hour │ Follow toward      │
│ busy-marker  │ Agent starts task │ On completion│ Avoid / pick other │
│ help-signal  │ Agent stuck >5min │ 25% per hour │ Assist if nearby   │
│ danger-flag  │ Error detected    │ 10% per hour │ Retreat & report   │
└──────────────┴───────────────────┴──────────────┴────────────────────┘

得: 映境標至沉件、衰率、應行之信號表。信號當簡、可組、獨義。

敗: 設感過繁→減為二:正(success trail)與負(danger flag)。多調可以吸斥引。基系行後加細。

三:定局互規

以局信息(己態+鄰信號)定各代循之簡規。

  1. 定代感半徑(何可感?)
  2. 按優先書 3-7 局規:
    • 規一(安):測 danger-flag→離
    • 規二(應):測 help-signal 且閒→趨
    • 規三(用):測 success-trail→趨最強
    • 規四(探):無信號→偏未探區隨動
    • 規五(沉):畢任→於處沉 success-trail
  3. 各規須:
    • :依只代可感
    • :一 if-then 句達
    • 無態(宜):無須代憶過態
  4. 心試:諸代循此規,所求群行生乎?

得: 各代獨行之優先規集。施於全群→生目標群行(foraging、construction、defense 等)。

敗: 心擬不生所求生行→規需饋環——代須可觀群動之果。加代群態之信號(如「任成率」)與按之調行之規。

四:校 quorum 感

立閾以代足同時觸群態變。

  1. 識須集同意之決(非僅個應):
    • 探轉用
    • 承新工地或棄舊
    • 自常升急應
  2. 各集決定:
    • Quorum threshold:須同意代之數或率
    • Sensing window:計信號之時段
    • Hysteresis:活與止之異閾(防震)
  3. 以信號積實 quorum:
    • 贊決之代沉 vote-signal
    • 窗內積票過 quorum 閾→決活
    • 票低於止閾→決反

得: 允群無首為集決之 quorum 閾。Hysteresis 隙防態速震。

敗: 群態震→廣 hysteresis 隙(如活於 70%、止於 30%)。群不達 quorum→降閾或廣窗。決過慢→縮窗,慎早共識。

五:測調生行

驗局規生所求群行,後調參。

  1. 以少代(5-10)行擬或試
  2. 觀:
    • 群收斂於意行乎?
    • 收斂幾久?
    • 任中態變時何?
    • 代敗或增時何?
  3. 調參:
    • 信號衰率:過速→無調憶;過緩→陳信號主
    • Quorum 閾:過低→早集決;過高→癱
    • 探用衡:過探→效低;過用→局優
  4. 壓測:
    • 忽去 30% 代→群復乎?
    • 倍代→仍調乎?
    • 引悖信號→解或死鎖?

得: 調參集,群自組趨目行、復擾、優擴。

敗: 群敗壓測→信號設過耦。簡之:減信號、增衰率(更鮮信息),保代無信號時有穩默行。零信號行理者勝依信號者。

  • 調問已分為識模(foraging、consensus、construction、defense、division of labor)
  • Stigmergic 信號表定,含沉件、衰率、代應
  • 局互規簡、局、優先(3-7 規)
  • Quorum 閾含 hysteresis 以防震
  • 小測顯生行合群目
  • 壓測(代去、加、信號擾)顯優雅降級

  • 過設信號:始過多信號型生惑。起於二信號(吸斥)、證要方加
  • 隱中思:「局規」求代知全態→非局。重構至各規依代直感
  • 忽衰:永不衰之信號生化石調態。各信號須合任時尺之半衰
  • 零 hysteresis:活與止無隙之 quorum 閾生態速震。止當低於活
  • 設同質:代異能→單規集敗。考角異規(見 scale-colony

  • forage-resources — 群調特於資搜及探用權衡
  • build-consensus — 分意機深探,擴此技之 quorum 感
  • defend-colony — 群禦模,築於此信號與規架
  • scale-colony — 群過初調設時之擴法
  • adapt-architecture — 架變 morphic 技,群調觸構變時補
  • deploy-to-kubernetes — 實分系部,群調模施於此
  • plan-capacity — 知於群擴動之容謀
  • coordinate-reasoning — AI 自施變;映 stigmergic 信號至脈管,含信息衰率與局協

GitHub Repository

pjt222/agent-almanac
Path: i18n/wenyan-ultra/skills/coordinate-swarm
0
agentsagentskillsai-assisted-developmentclaude-codeskillsteams

Related Skills

content-collections

Meta

This skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.

View skill

polymarket

Meta

This skill enables developers to build applications with the Polymarket prediction markets platform, including API integration for trading and market data. It also provides real-time data streaming via WebSocket to monitor live trades and market activity. Use it for implementing trading strategies or creating tools that process live market updates.

View skill

creating-opencode-plugins

Meta

This skill helps developers create OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It provides the plugin structure, event API specifications, and implementation patterns for JavaScript/TypeScript modules. Use it when you need to intercept, monitor, or extend the OpenCode AI assistant's lifecycle with custom event-driven logic.

View skill

sglang

Meta

SGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.

View skill