广告合作
MCP.so
登录
A

Agency MCP Server

@npupko

代码仓库

基本信息

分类

其他

传输方式

stdio

作者

npupko

提交者

Никита Пупко

关于 Agency MCP Server

暂无概览

Agency MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "agency": {
      "command": "npx",
      "args": [
        "-y",
        "agency-mcp-server"
      ]
    }
  }
}

Agency MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。

点击上方按钮,在临时沙箱中运行此服务器并实时获取工具列表。

概览

What is Agency Mcp Server?

Agency Mcp Server gives AI assistants instant access to 150+ expert agent templates (e.g., game economy designer, security auditor, technical writer) through the Model Context Protocol. It is designed for users of Claude Code, Cursor, Windsurf, and other MCP clients who need specialist help without loading every agent into conversation context.

How to use Agency Mcp Server?

Install via Claude Code plugin (/plugin add, /plugin install), CLI (claude mcp add agency -- npx -y agency-mcp-server), or add the MCP config entry shown in the README. On first launch, templates auto-clone to ~/.cache/agency-mcp-server/ and update every 24 hours.

Key features of Agency Mcp Server

  • 150+ specialist agent templates available on demand
  • Context-efficient: idle uses ~55 tokens instead of ~8,300
  • Auto‑updating templates with configurable update interval
  • Two tools: agency_search and agency_browse
  • Supports custom local templates or a custom Git repository

Use cases of Agency Mcp Server

  • Getting expert‑quality help for a specific task (e.g., security audit, game economy design)
  • Reducing context overhead when using many agents in Claude Code or similar tools
  • Browsing available specialist agents by division (Engineering, Design, Game Development, etc.)
  • Using your own curated agent templates while still benefiting from the MCP interface

常见问题

1

How does the context cost compare to installing all agents locally?

With 160+ agents installed locally, every conversation uses ~8,300 tokens. With the MCP server idle, it uses ~55 tokens; searching adds ~350 tokens; using one agent adds ~2,700 tokens median.

2

Can I use my own agent templates?

Yes. Set the `AGENCY_AGENTS_PATH` environment variable to point at a local directory, or set `AGENCY_REPO_URL` to a custom Git repository.

3

How do I verify the server is working?

Ask your assistant to search for an agent, e.g., "Search for a game economy designer agent." You should see results from the `agency_search` tool. On first run, templates auto‑download (~30 seconds).

4

What tools does the server provide?

It provides `agency_search(query, division?)` to find agents by task description, and `agency_browse(division?)` to list divisions or agents within a division.

5

Can I disable automatic template updates?

Yes. Set the `AGENCY_AUTO_UPDATE` environment variable to `false`. You can also change the update interval with `AGENCY_UPDATE_INTERVAL` (default 24 hours).

评论

其他 分类下的更多 MCP 服务器

P

PickFu MCP

PickFu MCP is a Model Context Protocol (MCP) server that connects AI assistants, developers, marketers, product managers, and AI agents to PickFu's consumer research platform. It enables users to create surveys, conduct

MCP Toolbox for Databases

googleapis

MCP Toolbox for Databases is an open source MCP server for databases.

15.9K

MaxKB

1Panel-dev

🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。

22.0K

Handl

Handl speaks the Model Context Protocol. Your whole billing operation — projects, invoices, clients, reports, approvals — as tools any AI assistant can call, inside your autonomy rules.

Klarix Intelligence Engine

Klarix

B2B competitive intelligence, live sales battlecards, 5-axis ICP prospect fit scoring, and tech stack teardowns over MCP and REST with cited public-web evidence.

广告合作