northseadl/norix-skills / llm-agent-dev
llm-agent-dev
UnverifiedLLM Agent engineering: pattern selection (12-mode matrix), data simulation, convergence iteration. Covers intent routing, function calling, ReAct, MCP, prompt chaining, guardrails, evaluation.
references/paradigms/pattern-matrix.md
View on GitHub# 模式矩阵全景 (Pattern Matrix Overview)
> 12 种 Agent 模式的特性对比与选择指南
## 模式光谱
```
简单 ◀──────────────────────────────────────────────────────▶ 复杂
① CREX ② Slot ③ FC ④ Workflow ⑨ Chain ⑩ Parallel
⑥ Router ⑦ MCP ⑧ Handoff ⑤ ReAct ⑪ Orch ⑫ Eval-Opt
确定性 ◀────────────────────────────────────────────────────▶ 涌现性
快速 ◀────────────────────────────────────────────────────▶ 慢速
可控 ◀────────────────────────────────────────────────────▶ 灵活
```
## 完整对比矩阵
| # | 模式 | 延迟 | 可预测性 | 灵活性 | LLM依赖 | 成本 | 最佳场景 |
|---|------|------|---------|--------|---------|------|---------|
| ① | CREX Rule Router | 50-500ms | ★★★★★ | ★★☆☆☆ | 可选 | 极低 | FAQ/查询/导航 |
| ② | Slot Filling | 100-800ms | ★★★★☆ | ★★★☆☆ | 可选 | 低 | 预约/下单/表单 |
| ③ | Function Calling | 500-2000ms | ★★★★☆ | ★★★★☆ | 是 | 中 | API工具调用 |
| ④ | Workflow Agent | 1-5s | ★★★★☆ | ★★★☆☆ | 部分 | 中 | 固定多步流程 |
| ⑤ | ReAct | 3-15s | ★★☆☆☆ | ★★★★★ | 核心 | 高 | 推理探索 |
| ⑥ | Router-Expert | 500-3000ms | ★★★☆☆ | ★★★★☆ | 路由可选 | 中 | 多领域系统 |
| ⑦ | MCP Tool Protocol | 取决于工具 | ★★★★☆ | ★★★★☆ | 是 | 中 | 工具生态 |
| ⑧ | Handoff/Delegation | 可变 | ★★★☆☆ | ★★★★☆ | 可选 | 低 | Agent->Agent/人工 |
| ⑨ | Prompt Chaining | 1-5s | ★★★★★ | ★★☆☆☆ | 是 | 中 | 顺序分解任务 |
| ⑩ | Parallelizing | max(子) | ★★★★☆ | ★★★☆☆ | 是 | 中-高 | 并行子任务 |
| ⑪ | Orchestrator-Worker | 2-10s | ★★★☆☆ | ★★★★★ | 核心 | 高 | 复杂度不确定 |
| ⑫ | Evaluator-Optimizer | 5-30s | ★★★★☆ | ★★★★☆ | 核心 | 高 | 质量迭代 |
## 模式间关系图
```
① CREX (基础循环)
╱ │ ╲
╱ │ ╲
② Slot ③ FC ⑥ Router
│ │ │
▼ ▼ ▼
④ Workflow ⑤ ReAct ⑧ Handoff
│ │ │
▼ ▼ ▼
⑨ Chain ⑩ Parallel ⑦ MCP
│ │
▼ ▼
⑪ Orchestrator-Worker
│
▼
⑫ Evaluator-Optimizer
```
**关系说明**:
- ①②③ 是原子模式,可独立使用
- ④⑤ 是 ①②③ 的增强(多步骤)
- ⑥⑧ 是跨域/跨 Agent 协调
- ⑨⑩⑪⑫ 是高级组合模式,通常组合使用原子模式
详细内容见各模式的独立参考文件: `patterns/{nn}-{name}.md`
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |