digihub-mcp

digihub-mcp

MCP server for the digihub.li content API, enabling Claude to read and write Stories, Solutions, and Sessions, and upload images for media.

Category
访问服务器

README

digihub-mcp

MCP server for the digihub.li content API. Plug it into Claude (Desktop or Code) and Claude can read, create, update and delete Stories, Solutions, Sessions, and upload images that get used as hero or inline media.


Requirements

What Where
1 Node.js 18+ (LTS recommended) https://nodejs.org/ — default installer is fine. Provides node, npm, npx.
2 An MCP-aware client Claude Desktop, Claude Code, or any other MCP client.
3 A digihub.li API token Ask the admin. Looks like `3
4 The base URL Usually https://digihub.li. The admin tells you if different.

To check Node.js is installed:

node --version   # should print v18.x or higher
npx --version    # should print 9.x or higher

If npx isn't found, install Node.js first. Claude Desktop won't bundle one for you.


Configure your Claude client

Same JSON for Desktop and Code — only the file path differs:

  • Desktop: claude_desktop_config.json (Windows: %APPDATA%\Claude\, macOS: ~/Library/Application Support/Claude/)
  • Code: .mcp.json in your project, or claude mcp add
{
  "mcpServers": {
    "digihub": {
      "command": "npx",
      "args": ["-y", "@stefan443/digihub-mcp@latest"],
      "env": {
        "DIGIHUB_BASE_URL": "https://digihub.li",
        "DIGIHUB_API_TOKEN": "PASTE_YOUR_TOKEN_HERE"
      }
    }
  }
}

Restart the client. First prompt: "Use the digihub whoami tool." You should see your name, email, and the abilities list.


Tools

Resource Tools
Stories list_stories, get_story, create_story, update_story, delete_story
Solutions list_solutions, get_solution, create_solution, update_solution, delete_solution
Sessions (events) list_sessions, get_session, create_session, update_session, delete_session
Media upload_media — pass file (local path) or url (server fetches it). Returns a /storage/... URL you put into hero_image or media[].url.
Self whoami

Required fields on create:

  • Story: slug, title.de/title.en, lead.de/lead.en, format (news, case_study, newsletter, use_case, success_story)
  • Solution: slug, title, summary
  • Session: slug, title, summary, format (free text), starts_at (ISO date)

Translations (title, lead, summary, body) are objects with de and/or en. At least one must be present. body accepts Markdown.

Filters on list_*: status, format, pillar_id, featured, per_page (≤ 100), page. list_sessions also takes upcoming: true.


Reference data

Pillar id key
AI Lab 1 lab
Coworking Hub 2 hub
Förderprogramme 3 fund

Tag slugs by type (pass tag_slugs: ["ai", "kmu"]):

  • business_area: lab, hub, fund
  • audience: kmu, verwaltung, startup, finanzdienstleister, industrie
  • topic: ai, datenschutz, …

The API only accepts existing tag slugs.


Example: a story with an image

// 1. Upload an image from a local path
{ "tool": "upload_media", "args": { "file": "C:/Pictures/hero.jpg", "subdir": "stories" } }
// → { "url": "/storage/stories/abc123.jpg" }

// 2. Create the story referencing it
{
  "tool": "create_story",
  "args": {
    "slug": "ki-workshop-im-mai",
    "pillar_id": 1,
    "title": { "de": "KI-Workshop im Mai", "en": "AI Workshop in May" },
    "lead": { "de": "Drei Stunden Hands-on mit echten KMU-Use-Cases." },
    "body": { "de": "# Worum geht's\n\nMarkdown wird gerendert ..." },
    "format": "news",
    "hero_image": "/storage/stories/abc123.jpg",
    "author_name": "digihub.li Team",
    "tag_slugs": ["ai", "kmu"],
    "status": "published",
    "published_at": "2026-05-01T08:00:00Z"
  }
}

upload_media also accepts { "url": "https://..." } — the server fetches the remote URL and re-hosts the file.


When something doesn't work

Error What's wrong
Server doesn't appear in Claude / logs say spawn npx ENOENT Node.js isn't installed or Claude can't find npx on PATH. Install Node.js, restart Claude. On macOS with Homebrew, use the absolute path (/opt/homebrew/bin/npx on Apple Silicon, /usr/local/bin/npx on Intel — which npx shows yours).
401 on whoami Token missing or revoked — ask admin for a new one
403 on any tool Your token doesn't have that ability — ask admin to widen it
422 on slug Slug must be lowercase letters, digits, dashes only; and unique
422 on tag_slugs Unknown tag — only use slugs from Reference data above
Image uploaded but doesn't show on the site Browser/CDN cache; hard-refresh or wait a few minutes

For anything else, send the admin the tool name, the arguments you used, and the error text.


MIT — see LICENSE.

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选