MCP.so
登录

Mcp Arcknowledge

@dragonjump

关于 Mcp Arcknowledge

This MCP server conveniently lets you configure and interact across all your custom webhook endpoints from a config(knowledgebase json).

基本信息

分类

其他

传输方式

stdio

发布者

dragonjump

提交者

sean sean

配置

暂无标准配置

该服务器的 README 中没有可解析的 MCP 配置块,请前往代码仓库查看安装说明。

代码仓库

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Mcp Arcknowledge?

Mcp Arcknowledge is a Model Context Protocol (MCP) server that lets you manage and query a list of custom knowledge base endpoints (webhooks). You register document source URLs, optionally with a description and API key, list all registered sources, and query them with a text question to get aggregated results.

How to use Mcp Arcknowledge?

Install Go, Python 3.6+, and UV, then clone the repository. Configure your endpoints in knowledge_document_sources.json (or set the DOCUMENT_SOURCES_PATH environment variable). Add the MCP configuration to your client (Claude Desktop, Cursor, or Cline) using the provided JSON snippet and restart the client. The server provides tools: add_new_knowledge_document_source, list_knowledge_document_sources, and query_knowledge_base.

Key features of Mcp Arcknowledge

  • Manage custom webhook endpoints as knowledge sources.
  • Register new document sources with URL, description, and optional API key.
  • List all registered knowledge sources with details.
  • Query sources using a text question.
  • Aggregate results from multiple sources in a single query.

Use cases of Mcp Arcknowledge

  • Connect multiple custom knowledge base endpoints into one unified MCP server.
  • Query internal company documentation hosted on custom webhook APIs.
  • Dynamically add new document sources and immediately search across them.

FAQ from Mcp Arcknowledge

What dependencies are required to run Mcp Arcknowledge?

Go, Python 3.6+, UV (Python package manager), and a compatible MCP client (Claude Desktop, Cursor, or Cline).

How do I configure the knowledge sources?

Edit the file knowledge_document_sources.json in the repository root, or set the DOCUMENT_SOURCES_PATH environment variable in your MCP configuration to point to a custom JSON file.

Where are the data sources stored?

All storage is runtime‑local inside the main Python server. No external database is used.

What transport does the server use?

The server uses standard MCP transport over stdin/stdout, compatible with Claude Desktop, Cursor, Cline, and Windsurf.

How do I add a new knowledge source?

Use the add_new_knowledge_document_source tool with a URL, an optional description, and an optional API key. Example: "Add new arcknowledge data source. Endpoint is http://something.com/api/123. Purpose is to handle questions on 123 topic. Api key is 'sk-2123123'."

评论

其他 分类下的更多 MCP 服务器