Communitygithub.com

claude-devfleet

透過 Claude DevFleet 協調多代理程式編碼任務 — 規劃專案、在隔離的工作樹中分派並行代理、監控進度並讀取結構化報告。適用於需要在隔離工作樹中分派並行編碼代理並追蹤其報告的情境。

claude-devfleet 是什麼?

這個技能讓你能夠協調多個 Claude Code 代理,在隔離的 git 工作樹中並行執行編碼任務。它使用一個獨立的 DevFleet 伺服器,透過 MCP 協定與 Claude 整合。主要工作流程包括:使用 `plan_project` 根據提示建立專案和任務有向無環圖、向用戶展示計劃並取得批准、使用 `dispatch_mission` 啟動代理、使用 `get_mission_status` 或 `get_dashboard` 監控進度,以及使用 `get_report` 取得結構化報告。DevFleet 預設最多支援 3 個並行代理,並會自動在依賴任務完成後觸發後續任務。這對於將大型工作分解為可並行處理的單元、加速開發週期,以及清楚地了解每個代理的貢獻(包括檔案變更、錯誤和後續步驟)非常有用。

相容平台✓Claude Code~Codex CLI~Cursor
npx skills add https://github.com/affaan-m/ECC/tree/main/skills/claude-devfleet

在你喜歡的 AI 中提問

開啟一個已預先載入此 Agent Skill 的新對話。

說明文件

Claude DevFleet 多代理協調

使用時機

當你需要分派多個 Claude Code 代理,讓它們在隔離的 git 工作樹中並行處理編碼任務時,請使用此技能。每個代理都擁有完整的工具集。

設定

DevFleet 伺服器是一個獨立的專案,並未與 ECC 捆綁。請先從其倉庫安裝並執行:https://github.com/LEC-AI/claude-devfleet

然後透過 MCP 連接到執行中的實例:

claude mcp add devfleet --transport http http://localhost:18801/mcp

首次使用前,請驗證監聽在端口 18801 上的程序確實是您安裝的 DevFleet 二進位檔(請參閱關於 localhost MCP 伺服器的 SECURITY.md)。

運作方式

用戶 → "使用認證和測試建立一個 REST API"
  ↓
plan_project(prompt) → project_id + 任務有向無環圖
  ↓
向用戶展示計劃 → 取得批准
  ↓
dispatch_mission(M1) → 代理 1 在工作樹中啟動
  ↓
M1 完成 → 自動合併 → 自動分派 M2(依賴於 M1)
  ↓
M2 完成 → 自動合併
  ↓
get_report(M2) → files_changed, what_done, errors, next_steps
  ↓
回報給用戶

工具

工具用途
plan_project(prompt)讓 AI 將一段描述分解成一個帶有連鎖任務的專案
create_project(name, path?, description?)手動建立一個專案,回傳 project_id
create_mission(project_id, title, prompt, depends_on?, auto_dispatch?)新增一個任務。depends_on 是任務 ID 字串的列表(例如 ["abc-123"])。設定 auto_dispatch=true 可在依賴項滿足時自動啟動。
dispatch_mission(mission_id, model?, max_turns?)啟動一個代理來執行任務
cancel_mission(mission_id)停止一個正在執行的代理
wait_for_mission(mission_id, timeout_seconds?)阻塞等待直到任務完成(見下方備註)
get_mission_status(mission_id)檢查任務進度,不阻塞
get_report(mission_id)讀取結構化報告(檔案變更、測試、錯誤、後續步驟)
get_dashboard()系統概覽:執行中的代理、統計資料、近期活動
list_projects()瀏覽所有專案
list_missions(project_id, status?)列出專案中的任務

關於 wait_for_mission 的備註: 這會阻塞對話,最多持續 timeout_seconds 秒(預設 600)。對於長時間執行的任務,建議每隔 30–60 秒輪詢 get_mission_status 來代替,以便用戶看到進度更新。

工作流程:計劃 → 分派 → 監控 → 報告

  1. 計劃:呼叫 plan_project(prompt="...") → 回傳 project_id 以及包含 depends_on 鏈和 auto_dispatch=true 的任務列表。
  2. 展示計劃:向用戶顯示任務標題、型別和依賴鏈。
  3. 分派:在根任務(depends_on 為空)上呼叫 dispatch_mission(mission_id=<第一個任務的ID>)。剩餘的任務會在其依賴項完成時自動分派(因為 plan_project 為它們設定了 auto_dispatch=true)。
  4. 監控:呼叫 get_mission_status(mission_id=...) 或 get_dashboard() 來檢查進度。
  5. 報告:當任務完成時,呼叫 get_report(mission_id=...)。向用戶分享重點摘要。

並行度

DevFleet 預設最多同時執行 3 個代理(可透過 DEVFLEET_MAX_AGENTS 配置)。當所有槽位滿載時,auto_dispatch=true 的任務會在工作監控器中排隊,並在槽位空出時自動分派。檢查 get_dashboard() 以了解當前槽位使用情況。

範例

全自動:計劃並啟動

  1. plan_project(prompt="...") → 顯示包含任務和依賴關係的計劃。
  2. 分派第一個任務(depends_on 為空的那個)。
  3. 剩餘任務會在其依賴項解析完成後自動分派(它們已設定 auto_dispatch=true)。
  4. 回報專案 ID 和任務數量,讓用戶知道啟動了什麼。
  5. 定期輪詢 get_mission_status 或 get_dashboard(),直到所有任務達到終止狀態(completed、failed 或 cancelled)。
  6. 對每個終止狀態的任務執行 get_report(mission_id=...) — 總結成功之處,並指出失敗及其錯誤和後續步驟。

手動:逐步控制

  1. create_project(name="我的專案") → 回傳 project_id。
  2. 為第一個(根)任務執行 create_mission(project_id=project_id, title="...", prompt="...", auto_dispatch=true) → 取得 root_mission_id。 為每個後續任務執行 create_mission(project_id=project_id, title="...", prompt="...", auto_dispatch=true, depends_on=["<root_mission_id>"])。
  3. 在第一個任務上執行 dispatch_mission(mission_id=...) 以啟動鏈條。
  4. 完成後執行 get_report(mission_id=...)。

順序執行並審查

  1. create_project(name="...") → 取得 project_id。
  2. create_mission(project_id=project_id, title="實作功能", prompt="...") → 取得 impl_mission_id。
  3. dispatch_mission(mission_id=impl_mission_id),然後輪詢 get_mission_status 直到完成。
  4. get_report(mission_id=impl_mission_id) 以審查結果。
  5. create_mission(project_id=project_id, title="審查", prompt="...", depends_on=[impl_mission_id], auto_dispatch=true) — 由於依賴項已滿足,會自動啟動。

指導原則

  • 在分派之前,務必先向用戶確認計劃,除非他們表示可以直接執行。
  • 報告狀態時,請包含任務標題和 ID。
  • 如果任務失敗,請在重試之前先讀取其報告。
  • 在大量分派之前,檢查 get_dashboard() 以了解代理槽位可用情況。
  • 任務依賴關係形成一個有向無環圖 — 不要建立循環依賴。
  • 每個代理在隔離的 git 工作樹中執行,並在完成時自動合併。如果發生合併衝突,變更會保留在代理的工作樹分支上,等待手動解決。
  • 手動建立任務時,如果希望它們在依賴項完成時自動觸發,請務必設定 auto_dispatch=true。如果沒有此標誌,任務將保持 draft 狀態。

Individual skills in this repo

This repo contains 20 individual skills — each has its own dedicated page.

accessibility

Design, implement, and audit inclusive digital products using WCAG 2.2 Level AA. Use when building or auditing UI that must meet WCAG 2.2 Level AA, or when reviewing a change for keyboard, contrast, or screen-reader support.

affaan-m/content-engine

Create platform-native content systems for X, LinkedIn, TikTok, YouTube, newsletters, and repurposed multi-platform campaigns. Use when the user wants social posts, threads, scripts, content calendars, or one source asset adapted cleanly across platforms.

affaan-m/ECC

Multi-platform content distribution across X, LinkedIn, Threads, and Bluesky. Adapts content per platform using content-engine patterns. Never posts identical content cross-platform. Use when the user wants to distribute content across social platforms.

affaan-m/fal-ai-media

Unified media generation via fal.ai MCP — image, video, and audio. Covers text-to-image (Nano Banana), text/image-to-video (Seedance, Kling, Veo 3), text-to-speech (CSM-1B), and video-to-audio (ThinkSound). Use when the user wants to generate images, videos, or audio with AI.

affaan-m/manim-video

日本語翻訳:このファイルは manim-video 用の日本語翻訳が必要です

affaan-m/remotion-video-creation

Remotion のベストプラクティス - React で動画を作成する。3D、アニメーション、音声、字幕、チャート、トランジションなどをカバーするドメイン固有の29のルール。

affaan-m/video-editing

AI-assisted video editing workflows for cutting, structuring, and augmenting real footage. Covers the full pipeline from raw capture through FFmpeg, Remotion, ElevenLabs, fal.ai, and final polish in Descript or CapCut. Use when the user wants to edit video, cut footage, create vlogs, or build video content.

agent-architecture-audit

Full-stack diagnostic for agent and LLM applications. Audits the 12-layer agent stack for wrapper regression, memory pollution, tool discipline failures, hidden repair loops, and rendering corruption. Produces severity-ranked findings with code-first fixes. Essential for developers building agent applications, autonomous loops, or any LLM-powered feature. Use when an agent or LLM feature misbehaves and the failing layer is unknown, or before shipping an agent stack.

agent-eval

Head-to-head comparison of coding agents (Claude Code, Aider, Codex, etc.) on custom tasks with pass rate, cost, time, and consistency metrics. Use when choosing between coding agents, or when a change to an agent setup needs measured pass rate, cost, and time rather than an impression.

agent-harness-construction

Design and optimize AI agent action spaces, tool definitions, and observation formatting for higher completion rates. Use when defining or revising an agent

agentic-engineering

Operate as an agentic engineer using eval-first execution, decomposition, and cost-aware model routing. Use when planning or executing engineering work that agents will carry out end to end.

agentic-os

Build persistent multi-agent operating systems on Claude Code. Covers kernel architecture, specialist agents, slash commands, file-based memory, scheduled automation, and state management without external databases. Use when building a persistent multi-agent system on Claude Code with its own memory, commands, and scheduling.

agent-introspection-debugging

Structured self-debugging workflow for AI agent failures using capture, diagnosis, contained recovery, and introspection reports. Use when an agent run fails and you need a reproducible diagnosis instead of a retry.

agent-payment-x402

Add x402 payment execution to AI agents with per-task budgets, spending controls, and non-custodial wallets. Supports Base through agentwallet-sdk and X Layer through OKX Payments / OKX Agent Payments Protocol. Use when an agent must pay for something itself and needs per-task budgets, spending controls, and a non-custodial wallet.

agent-self-evaluation

Use after completing any non-trivial task. The agent self-rates its output on 5 axes — accuracy, completeness, clarity, actionability, conciseness — with concrete evidence per criterion. Produces a structured 1-5 scorecard with specific improvement suggestions.

agent-sort

Build an evidence-backed ECC install plan for a specific repo by sorting skills, commands, rules, hooks, and extras into DAILY vs LIBRARY buckets using parallel repo-aware review passes. Use when ECC should be trimmed to what a project actually needs instead of loading the full bundle.

ai-first-engineering

Engineering operating model for teams where AI agents generate a large share of implementation output. Use when setting team process, review gates, or ownership rules for a codebase largely written by agents.

ai-regression-testing

Regression testing strategies for AI-assisted development. Sandbox-mode API testing without database dependencies, automated bug-check workflows, and patterns to catch AI blind spots where the same model writes and reviews code. Use when adding regression coverage to AI-assisted code, or when the same model both wrote and reviewed a change.

android-clean-architecture

Clean Architecture patterns for Android and Kotlin Multiplatform projects — module structure, dependency rules, UseCases, Repositories, and data layer patterns. Use when structuring modules, layers, or data flow in an Android or KMP project.

angular-developer

Generates Angular code and provides architectural guidance. Trigger when creating projects, components, or services, or for best practices on reactivity (signals, linkedSignal, resource), forms, dependency injection, routing, SSR, accessibility (ARIA), animations, styling (component styles, Tailwind CSS), testing, or CLI tooling.

相關技能