What problem does MCP Context Manager solve compared to plain context windows?
It uses knowledge graphs to free up the AI’s context window (performance) and minimize token cost (efficiency), providing persistent, organized context across sessions.
@tejpalvirk
基本信息
MCP Servers for persistent context across work sessions using knowledge graphs
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"contextmanager": {
"command": "npx",
"args": [
"github:tejpalvirk/contextmanager"
]
}
}
}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。
MCP Context Manager is a collection of Model Context Protocol (MCP) servers that provide AI models with persistent context across work sessions throughout the project lifecycle. It stores context in domain‑specific knowledge graphs and offers a central manager for unified access to multiple domain servers.
Install via npx github:tejpalvirk/contextmanager (recommended) or globally with npm install -g github:tejpalvirk/contextmanager and run mcp-server-contextmanager. Use tools such as setActiveDomain, startsession, endsession, buildcontext, loadcontext, and deletecontext to manage context. Configure domain selection with the --domains flag and data paths with environment variables MEMORY_FILE_PATH and SESSIONS_FILE_PATH.
It uses knowledge graphs to free up the AI’s context window (performance) and minimize token cost (efficiency), providing persistent, organized context across sessions.
Node.js v16 or higher and npm v7 or higher are required. No external databases or services are needed beyond the local file system.
Data is stored in two files per domain: a memory file (default `<domain_directory>/memory.json`) and a sessions file (default `<domain_directory>/sessions.json`). Paths can be customized with environment variables `MEMORY_FILE_PATH` and `SESSIONS_FILE_PATH`.
It uses the MCP Client SDK to communicate between the Context Manager and domain servers, and it exposes an MCP server interface for integration with clients like Claude Desktop.
Use the `--port` argument to specify a different port (default is 3000), for example `npx github:tejpalvirk/contextmanager --port 3001`.
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

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.
Hindsight: Agent Memory That Learns
YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context
MCP Toolbox for Databases is an open source MCP server for databases.
评论