KS
Killer-Skills

workflow-multi-cli-plan — Categories.community claude, claude-code, cli-tools, codex, context-management, gemini-cli, json-driven, qwen-code, task-orchestration, workflow-automation

v1.0.0
GitHub

About this Skill

Ideal for Advanced AI Agents like AutoGPT and LangChain requiring unified multi-CLI collaborative planning and execution capabilities. JSON-driven multi-agent cadence-team development framework with intelligent CLI orchestration (Gemini/Qwen/Codex), context-first architecture, and automated workflow execution

# Core Topics

catlog22 catlog22
[1.4k]
[113]
Updated: 3/2/2026

Quality Score

Top 5%
61
Excellent
Based on code quality & docs
Installation
SYS Universal Install (Auto-Detect)
Cursor IDE Windsurf IDE VS Code IDE
> npx killer-skills add catlog22/Claude-Code-Workflow/workflow-multi-cli-plan

Agent Capability Analysis

The workflow-multi-cli-plan MCP Server by catlog22 is an open-source Categories.community integration for Claude and other AI agents, enabling seamless task automation and capability expansion. Optimized for claude, claude-code, cli-tools.

Ideal Agent Persona

Ideal for Advanced AI Agents like AutoGPT and LangChain requiring unified multi-CLI collaborative planning and execution capabilities.

Core Value

Empowers agents to orchestrate JSON-driven workflows with intelligent CLI tools like Gemini, Qwen, and Codex, leveraging context-first architecture for automated execution via protocols like multi-cli-plan and lite-execute.

Capabilities Granted for workflow-multi-cli-plan MCP Server

Automating complex workflow executions across multiple CLIs
Generating collaborative plans through multi-CLI discussions
Optimizing workflow cadence with context-aware team development frameworks

! Prerequisites & Limits

  • Requires JSON-driven workflow definitions
  • Dependent on compatible CLI tools like Gemini, Qwen, and Codex
  • Internal handoff to lite-execute for Phase 2 execution
SKILL.md
Readonly

Workflow Multi-CLI Plan

Unified multi-CLI collaborative planning and execution skill. Routes to multi-cli-plan (ACE context + multi-CLI discussion + plan generation) which then hands off to lite-execute (Phase 2) internally for execution.

Architecture Overview

┌─────────────────────────────────────────────────────────┐
│  SKILL.md (Router + Prompt Enhancement)                  │
│  → Enhance prompt → Dispatch to Phase 1                 │
└──────────────────────┬──────────────────────────────────┘
                       │
                       ↓
     ┌──────────────┐        ┌───────────┐
     │multi-cli-plan│        │lite-execute│
     │   Phase 1    │─handoff→│  Phase 2   │
     │  Plan+Exec   │        │ (internal) │
     └──────────────┘        └───────────┘

Mode Detection & Routing

javascript
1const args = $ARGUMENTS 2const mode = 'plan' // workflow-multi-cli-plan always starts with planning

Routing Table:

TriggerModePhase DocumentDescription
workflow-multi-cli-planplanphases/01-multi-cli-plan.mdMulti-CLI collaborative planning (ACE context → discussion → plan → execute)

Interactive Preference Collection

Before dispatching, collect workflow preferences via AskUserQuestion:

javascript
1// ★ 统一 auto mode 检测:-y/--yes 从 $ARGUMENTS 或 ccw 传播 2const autoYes = /\b(-y|--yes)\b/.test($ARGUMENTS) 3 4if (autoYes) { 5 // 自动模式:跳过所有询问,使用默认值 6 workflowPreferences = { autoYes: true } 7} else if (mode === 'plan') { 8 const prefResponse = AskUserQuestion({ 9 questions: [ 10 { 11 question: "是否跳过所有确认步骤(自动模式)?", 12 header: "Auto Mode", 13 multiSelect: false, 14 options: [ 15 { label: "Interactive (Recommended)", description: "交互模式,包含确认步骤" }, 16 { label: "Auto", description: "跳过所有确认,自动执行" } 17 ] 18 } 19 ] 20 }) 21 workflowPreferences = { 22 autoYes: prefResponse.autoMode === 'Auto' 23 } 24}

workflowPreferences is passed to phase execution as context variable, referenced as workflowPreferences.autoYes within phases.

Prompt Enhancement

After collecting preferences, enhance context and dispatch:

javascript
1// Step 1: Load project context via ccw spec 2Bash('ccw spec load --category planning') 3 4// Step 2: Log available context 5console.log('Project context loaded via: ccw spec load --category planning') 6 7// Step 3: Dispatch to Phase 1 (workflowPreferences available as context) 8// Read phases/01-multi-cli-plan.md and execute 9// Phase 1 internally hands off to Phase 2 (lite-execute) after plan approval

Compact Recovery (Phase Persistence)

Multi-phase execution (multi-cli-plan → lite-execute) spans long conversations. Uses 双重保险: TodoWrite 跟踪 active phase 保护其不被压缩,sentinel 作为兜底。

⚠️ COMPACT DIRECTIVE: Context compression MUST check TodoWrite phase status. The phase currently marked in_progress is the active execution phase — preserve its FULL content. Only compress phases marked completed or pending.

Execution Flow

Plan Mode (workflow-multi-cli-plan)

1. Collect preferences via AskUserQuestion (autoYes)
2. Enhance prompt with project context availability
3. Read phases/01-multi-cli-plan.md
4. Execute multi-cli-plan pipeline (Phase 1-5 within the phase doc)
5. Phase 5 directly reads and executes Phase 2 (lite-execute) with executionContext

Usage

Task description provided as arguments → interactive preference collection → multi-CLI planning pipeline → internal execution handoff to Phase 2 (lite-execute).

Phase Reference Documents

PhaseDocumentPurposeCompact
1phases/01-multi-cli-plan.mdComplete multi-CLI planning pipeline: ACE context, iterative discussion, options, user decision, plan generation, handoffTodoWrite 驱动
2phases/02-lite-execute.mdComplete execution engine: input modes, task grouping, batch execution, code reviewTodoWrite 驱动 + 🔄 sentinel

Compact Rules:

  1. TodoWrite in_progress → 保留完整内容,禁止压缩
  2. TodoWrite completed → 可压缩为摘要
  3. 🔄 sentinel fallback → Phase 2 包含 compact sentinel;若 compact 后仅存 sentinel 而无完整 Step 协议,必须立即 Read("phases/02-lite-execute.md") 恢复

Related Skills

Looking for an alternative to workflow-multi-cli-plan or building a Categories.community AI Agent? Explore these related open-source MCP Servers.

View All

widget-generator

Logo of f
f

widget-generator is an open-source AI agent skill for creating widget plugins that are injected into prompt feeds on prompts.chat. It supports two rendering modes: standard prompt widgets using default PromptCard styling and custom render widgets built as full React components.

149.6k
0
Design

chat-sdk

Logo of lobehub
lobehub

chat-sdk is a unified TypeScript SDK for building chat bots across multiple platforms, providing a single interface for deploying bot logic.

73.0k
0
Communication

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication