本文へ移動
cccskills
無料GitHub で公開

workflow-automator

Takes a manual business workflow description and designs the automated version. Maps current steps, handoffs, decision points, and bottlenecks. Designs automated flow with triggers, conditions, actions, and error handling. Outputs workflow-automation.md with before/after Mermaid diagrams, tool recommendations, implementation steps, and time savings estimate.

インストール方法を見る

含まれるファイル(7)

  • SKILL.md3.8 KB
  • references/analysis-framework.md2.1 KB
  • references/automation-design.md3.4 KB
  • references/diagram-and-estimation-standards.md1.5 KB
  • references/intake.md1.1 KB
  • references/output-template.md7.6 KB
  • references/tool-recommendations.md3.2 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Workflow Automator

Transform a manual business workflow into an optimized automated system: map how work gets done today, then design a complete automated replacement with triggers, conditions, actions, branching, and error handling. Deliver a comprehensive workflow-automation.md.

Contents

  • references/intake.md -- What to gather and how to ask for missing detail
  • references/analysis-framework.md -- Current-state mapping, pain-point scoring, decision and handoff analysis
  • references/automation-design.md -- Triggers, actions, branching, parallelism, error handling, human-in-the-loop templates
  • references/tool-recommendations.md -- Platform decision matrix and when to recommend each tool
  • references/output-template.md -- Full workflow-automation.md structure to fill in
  • references/diagram-and-estimation-standards.md -- Mermaid conventions and time/ROI estimation rules

Workflow

  1. Intake. Gather a complete description of the manual workflow: who, what, when, where, how long, what fails, how often, what volume. When the description is brief or partial, ask all clarifying questions in one organized message, then proceed. See references/intake.md.
  2. Map the current state. Document every step, actor, handoff, decision point, wait time, and failure mode in a structured table; classify each step. See references/analysis-framework.md.
  3. Identify pain points. Score each step on automation potential, impact, and risk. Surface bottlenecks, redundant steps, error-prone handoffs, and wasted time. See references/analysis-framework.md.
  4. Design the automated flow. Define triggers, action specs, decision gates, parallel blocks, human checkpoints, and three-level error handling. See references/automation-design.md.
  5. Recommend tools. Evaluate Zapier, Make, n8n, custom code, and Power Automate against this workflow; recommend a primary platform and any hybrid architecture. See references/tool-recommendations.md.
  6. Estimate impact. Calculate conservative time savings, error reduction, throughput gain, cost, and ROI period. See references/diagram-and-estimation-standards.md.
  7. Deliver. Generate the full workflow-automation.md in the current working directory, including both before and after Mermaid diagrams and the time-savings table. See references/output-template.md and references/diagram-and-estimation-standards.md.

Output Rules

  • Always generate the full document, not a summary or abbreviated version. Make it self-contained enough to implement from alone.
  • Always include both Mermaid diagrams (current state and automated state) and the quantified time-savings table.
  • Use no emojis anywhere in the output.

Quality Checklist

Before delivering, verify:

  • Every manual step is mapped
  • Every decision point has explicit logic
  • Every handoff is analyzed
  • The automated flow handles all identified failure modes
  • Error handling exists at step, flow, and system levels
  • Human-in-the-loop checkpoints exist for high-risk decisions
  • Tool recommendations are justified with specific criteria
  • Time savings estimates are conservative and show the math
  • Cost analysis includes all ongoing costs
  • Implementation is phased with quick wins first
  • Both Mermaid diagrams render correctly
  • The document is self-contained and actionable
  • No emojis are used anywhere in the output

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Audit websites for accessibility issues and WCAG compliance. Use when checking accessibility, fixing a11y issues, or ensuring WCAG compliance.

日本語の概要は準備中です。原文の説明を表示しています。

OneWave-AI/claude-skills3392026年10月2日 更新

Deploy a 2-layer parallel agent hierarchy for large, parallelizable work — big refactors, multi-file migrations, codebase-wide audits, bulk generation. A top-tier commander (Fable or Opus) orchestrates the swarms; the user picks a power level (Max Power / Heavy / Balanced / Economy) that sets the Opus/Sonnet/Haiku model mix per layer. Layer 1 is 3-50+ specialist agents, each with its own full context window; Layer 2 is 2+ sub-agents per member. Includes git safety, tiered sizing, a pre-deploy gate, phantom-completion checks, and multi-wave follow-up.

日本語の概要は準備中です。原文の説明を表示しています。

OneWave-AI/claude-skills3392026年10月2日 更新

Use to test whether an AI agent, prompt, or skill actually works before or after it ships — build a small eval set, score runs, and catch regressions when the prompt or model changes. Trigger when the user asks whether a prompt is good, wants to compare models or versions, sees inconsistent agent output, or is about to put an agent in front of customers.

日本語の概要は準備中です。原文の説明を表示しています。

OneWave-AI/claude-skills3392026年10月2日 更新

Deploys swarms of sub-agents for massive parallel data processing tasks. Unlike agent-army (which is for code changes), this is for DATA tasks -- processing 1000 documents, analyzing datasets, bulk content generation. Configurable swarm size, task distribution, result aggregation, progress tracking, and error recovery.

日本語の概要は準備中です。原文の説明を表示しています。

OneWave-AI/claude-skills3392026年10月2日 更新

Designs and deploys custom agent teams for specific business workflows. Interactive discovery of business processes, then generates complete team configurations with specialized agent roles, tool access, communication protocols, and handoff rules.

日本語の概要は準備中です。原文の説明を表示しています。

OneWave-AI/claude-skills3392026年10月2日 更新

Agent-to-Agent (A2A) communication protocol. Connect two or more Claude agents that pass messages, share context, delegate tasks, and collaborate. Implements structured handoffs, shared memory, and multi-agent conversations.

日本語の概要は準備中です。原文の説明を表示しています。

OneWave-AI/claude-skills3392026年10月2日 更新

OneWave-AI のスキルをすべて見る

このスキルの問題を報告する