菜单
Library of proven Make.com AI agent patterns. Returns the matching pattern blueprint — Mermaid diagram, module sequence, and wiring notes — based on the task type identified by ai-agent-designer. Called during System Design phase before plan-builder.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/cardinalconseils/make-automation-specialist --skill agent-pattern-library命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Designs AI agent architecture for Make.com scenarios. Determines model selection, tool inventory, memory strategy, looping pattern, and output contract. Called by scenario-orchestrator during System Design phase when ai_required is true, or directly by ai-agent-builder agent.
Pre-push blueprint JSON review for Make.com scenarios. Checks structural integrity, connection IDs, mapper references, filter logic, and error handler coverage before any scenario_update call.
Fetches exact Make.com module documentation before any blueprint is designed. Returns precise field names, required vs optional parameters, data types, and enumerated values. Eliminates improvisation and ensures first-shot accuracy.
基于 SOC 职业分类
正在显示 SKILL.md
| name | agent-pattern-library |
| description | Library of proven Make.com AI agent patterns. Returns the matching pattern blueprint — Mermaid diagram, module sequence, and wiring notes — based on the task type identified by ai-agent-designer. Called during System Design phase before plan-builder. |
Provides battle-tested Make.com AI agent patterns.
Do not design from scratch when a proven pattern exists. Match the task type to a pattern, load it, and adapt the specifics.
PATTERNS-BASIC.md — Patterns 1–4 (Classify-and-Route, Generator, Chunk-and-Summarize, Structured Extractor)PATTERNS-ADVANCED.md — Patterns 5–7 (RAG, ReAct Loop, Batch Processor) + selection logicPATTERNS-OUTPUT.md — Output contract returned to calling skill/agent| Task type | Pattern | Complexity | Ops/run estimate |
|---|---|---|---|
| Classifier / Router | Pattern 1: Classify-and-Route | Low | 3–5 |
| Generator | Pattern 2: Single-Shot Generator | Low | 2–3 |
| Summarizer | Pattern 3: Chunk-and-Summarize | Medium | 4–8 |
| Extractor | Pattern 4: Structured Extractor | Low | 2–3 |
| RAG (search + answer) | Pattern 5: Retrieval-Augmented Generator | Medium | 5–10 |
| ReAct agent | Pattern 6: ReAct Loop | High | 10–50+ |
| Batch processor | Pattern 7: Batch AI Processor | Medium | 3 × batch size |
IF task = "decide which category" OR "route to team" OR "score"
→ Pattern 1: Classify-and-Route
ELSE IF task = "write" OR "generate" OR "draft" OR "suggest"
→ Pattern 2: Single-Shot Generator
ELSE IF task = "summarize long document" OR "condense"
→ Pattern 3: Chunk-and-Summarize
ELSE IF task = "extract fields" OR "parse" OR "pull data from"
→ Pattern 4: Structured Extractor
ELSE IF task = "answer questions from" OR "FAQ" OR "knowledge base"
→ Pattern 5: RAG
ELSE IF task = "autonomously complete" OR "research and do" OR "multi-step"
→ Pattern 6: ReAct Loop
ELSE IF task = "process a list" OR "batch" OR "run on each row"
→ Pattern 7: Batch Processor