Files
nexus/wiki/entities/The-Agency.md

77 lines
3.1 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# The Agency
## Definition
The Agency 是一个由 AI 编码代理agent驱动的多智能体开发系统提供 **147 个专业化 Agent** 跨 **12 个业务部门**运作,涵盖 Engineering、Design、Finance、Game Dev、Marketing、Paid Media、Product、Project Management、Testing、Support、Spatial Computing、Specialized 等领域。
## Organizational Structure
```
The Agency
├── Engineering工程部门
├── Design设计部门
│ ├── UI Designer
│ ├── Brand Guardian
│ ├── Visual Storyteller
│ ├── UX Researcher
│ ├── Image Prompt Engineer
│ ├── Inclusive Visuals Specialist
│ └── Whimsy Injector
├── Finance财务部门
├── Game Dev游戏开发部门
├── Marketing市场营销部门
├── Paid Media付费媒体部门
├── Product产品部门
├── Project Management项目管理部门
├── Testing测试部门
├── Support支持部门
├── Spatial Computing空间计算部门
│ ├── XR Interface Architect
│ ├── XR Immersive Developer
│ ├── XR Cockpit Interaction Specialist
│ ├── visionOS Spatial Engineer
│ └── macOS Spatial/Metal Engineer
└── Specialized专业部门
├── Study Abroad Advisor
├── Agents Orchestrator
├── Identity Graph Operator
├── MCP Builder Agent
├── Agentic Identity & Trust Architect
├── Civil Engineer
├── Document Generator
├── Government Digital Presales Consultant
├── Healthcare Marketing Compliance Specialist
├── Compliance Auditor
├── Workflow Architect
├── Corporate Training Designer
├── Cultural Intelligence Strategist
├── Model QA Specialist
└── LSP/Index Engineer
```
## Core Design Principles
1. **鲜明性格**(拒绝通用人设)——每个 Agent 拥有明确的专业边界和独特性格
2. **明确交付物**(真实代码/模板)——每个 Agent 的输出可量化、可验证
3. **可量化指标**——每个 Agent 定义明确的成功指标
4. **经过验证的工作流**——Agent 行为遵循经过测试的工作流程
5. **学习记忆机制**——Agent 具备上下文记忆和自我改进能力
## Contribution Model
The Agency 接受外部贡献,核心方式:
- 创建全新智能体8 大分类)
- 优化现有智能体
- 分享成功案例
- 反馈问题
详见 [[contributing_zh-cn]] 和 [[contributing]](英文原版)。
## Relationship to Multi-Agent Reliability
The Agency 是 [[multi-agent-system-reliability]] 理论的最佳实践场域——通过 Agent 间协作模式Orchestration、Consensus Voting、Adversarial Debate、Knock-out实现系统级可靠性。[[identity-graph-operator]] 解决 Agent 间身份一致性问题,[[agentic-identity-trust]] 解决 Agent 间信任与授权问题,共同构成完整的多智能体基础设施。
## Aliases
- The Agency Multi-Agent System
- Agency Agents