jnMetaCode/agency-agents-zh

🎭 277 个即插即用的 AI 专家角色 — 支持 Claude Code/Cursor/Copilot 等 20 种工具，覆盖工程/设计/营销/金融等 20 个部门。含 64 个中国市场原创智能体（小红书/抖音/微信/飞书/钉钉/Qt 上位机/机械设计）。搭配编排器 agency-orchestrator，一句话即可让多位专家按 DAG 自动协作。

What it solves

This project provides a comprehensive library of 277 plug-and-play AI expert roles (agents) designed to move beyond generic prompts. It solves the problem of AI producing vague results by providing each agent with a specific identity, professional workflow, and defined deliverables, allowing users to simulate a full professional team (from C-suite executives to specialized engineers and marketers) within their AI tools.

How it works

The library consists of detailed Markdown files that define the "soul" and operational rules of various experts. These can be integrated into AI programming tools (like Claude Code, Cursor, Copilot, etc.) via installation scripts. Once installed, users can activate a specific expert mode using natural language (e.g., "Activate frontend developer mode"). The project also supports an orchestrator tool for multi-agent collaboration, allowing different experts to work together on a single task.

Who it’s for

  • AI-assisted developers using tools like Cursor or Claude Code who need specialized technical roles.
  • Solopreneurs wanting to simulate a corporate structure (CEO, CTO, CMO) to refine business strategies.
  • Digital marketers targeting the Chinese market (with specialized agents for Xiaohongshu, Douyin, WeChat, and Bilibili).
  • Enterprise users looking to standardize AI agent personas for their teams.

Highlights

  • Massive Role Library: 277 experts across 20 departments, including 64 original roles specifically tailored for the Chinese market.
  • Wide Tool Compatibility: Supports 20 different AI programming tools and CLI environments.
  • Structured Personas: Each agent includes a specific identity, key rules, and a professional delivery format rather than simple prompt templates.
  • Multi-Agent Orchestration: Compatible with the Agency Orchestrator for automated team-based workflows.
  • Vertical Domain Expertise: Includes niche roles such as Industrial PC Host Engineers, Mechanical Design Engineers, and Cross-border E-commerce operators.

Related

  • Project
    kangarooking/kangarooking-skills

    A collection of structured, open-source Agent Skills compatible with the Agent Skills standard, providing reusable workflows for content creation, image generation, and AI project engineering.

  • Project
    AgentiLoop/Agent

    AgentiLoop Agent is a native macOS app that lets any LLM act autonomously on your computer—coding, building, automating apps, running shell commands, fetching the web, sending iMessages, and responding to voice. It supports 23 providers, offers a self‑verifying task loop, sub‑agents, custom MCP tools, and a Swift‑based scripting system, all distributed as a signed, notarized binary.

  • Project
    vstorm-co/full-stack-ai-agent-template

    A CLI generator that scaffolds a production‑ready FastAPI + Next.js web app with built‑in AI‑agent runtimes, RAG pipelines, authentication, billing, observability and Docker/K8s deployment tooling.

  • Project
    jnMetaCode/agency-orchestrator

    A multi-agent orchestration tool that turns a single sentence into a coordinated team of specialized AI roles, running parallel DAG workflows with quality checks, supporting many LLM providers (including key-free reuse of existing coding CLIs), and offering both CLI and visual web interfaces.

  • Dispatch
    AI & Frontier Tech Roundup – Agentic AI, Large Model Inference, and Robotics Highlights

    Recent posts reveal a surge in agentic AI deployments, new open‑source inference tools that run frontier models on consumer hardware, and China’s expanding humanoid robot competitions.