MAM MCP Context Service
Enables querying context about the MAM product components and architecture through MCP, allowing agents to retrieve targeted product information on demand instead of loading full documentation.
README
MAM MCP Context Service
An MCP (Model Context Protocol) server that provides queryable context about the Mobile Access Management (MAM) product — the MAM Server, Ground Control Launchpad, and GroundControl Locker components, their integrations, and overall architecture.
Why
The MAM product is highly technical and deeply integrated across components and third-party/OS-level hooks. The existing MAM_KT_2 mkdocs knowledge base captures this well, but is too heavyweight for an agent to consume directly during an agentic development session — the only current option is indexing the entire doc tree into local memory, which is expensive and doesn't scale.
This service uses the OKF framework as a structural reference to build a metadata/service layer over the MAM_KT_2 knowledge base, exposing it through MCP so agentic tools (initially Claude) can query targeted product context on demand instead of loading the full doc tree.
See brief-mam-mcp-server.md for the product brief.
Setup
This service reads content from the MAM_KT_2 knowledge base at runtime. Symlink your local MAM_KT_2 checkout into the project root (the symlink is gitignored):
ln -s /path/to/MAM_KT_2 MAM_KT_2
No environment variables are required if the symlink is present. To point at a
different location instead, set MAM_KT_2_ROOT, or override MAM_SERVER_DETAIL_DIR
and MAM_SERVER_OVERVIEW_DIR directly.
Running the server
Install the package (in a virtualenv):
pip install -e ".[dev]"
Run the MCP server over stdio:
python -m mam_mcp
Point an MCP-aware client (e.g. Claude) at this command as a stdio server. The server exposes a single query tool backed by retrieval over the MAM Server documentation corpus — see brief-mam-mcp-server.md for scope.
Status
story-001 (single-component query, MAM Server) shipped. See slices/ for what's next.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。