広告掲載
MCP.so
ログイン

Strapi MCP Server

@l33tdawg

リポジトリ

基本情報

カテゴリ

その他

ライセンス

MIT

トランスポート

stdio

作者

l33tdawg

Strapi MCP Server について

An MCP server for your Strapi CMS

Strapi MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "strapi-mcp": {
      "command": "node",
      "args": [
        "--env-file=.env",
        "build/index.js"
      ]
    }
  }
}

Strapi MCP Server のツール

18

List all available content types in Strapi

Get entries for a specific content type with optional filtering, pagination, sorting, and population of relations

Get a specific entry by ID

Create a new entry for a content type

Update an existing entry

Delete an entry

Upload a media file to Strapi (max ~750KB file due to base64 context limits)

Upload a media file from local file path (max 10MB, avoids context overflow)

Get the schema (fields, types, relations) for a specific content type.

Connect related entries to an entry's relation field.

Disconnect related entries from an entry's relation field.

Create a new content type using the Content-Type Builder API (Requires Admin privileges).

Publish a specific entry.

Unpublish a specific entry.

List all available components in Strapi.

Get the schema for a specific component.

Create a new component.

Update an existing component.

概要

What is Strapi MCP?

Strapi MCP is a Model Context Protocol server that integrates with any Strapi CMS instance. It provides AI assistants with access to Strapi content types and entries, enabling automated content management through MCP resources and tools.

How to use Strapi MCP?

Install from npm (npm install strapi-mcp) or from source. Configure environment variables (STRAPI_URL, STRAPI_ADMIN_EMAIL, STRAPI_ADMIN_PASSWORD, and optional STRAPI_API_TOKEN and STRAPI_DEV_MODE). Run via Cursor MCP configuration (in ~/.cursor/mcp.json) or Claude Desktop config, or directly with node --env-file=.env build/index.js. Use built-in tools to manage Strapi content.

Key features of Strapi MCP

  • List and read content types as resources
  • Create, read, update, and delete entries
  • Upload media files via base64 or local file path
  • Connect and disconnect relations between entries
  • Manage components (list, create, update, get schema)
  • Publish and unpublish entries
  • Robust error handling and configuration validation

Use cases of Strapi MCP

  • Let AI assistants read and modify Strapi content types and entries
  • Automate content creation, updates, and publishing workflows
  • Manage component schemas and relationships programmatically

よくある質問

1

What authentication methods are supported?

You can use admin email/password (recommended for full schema access) or an API token as a fallback. The server prefers admin credentials and validates configuration to reject common placeholders.

2

What environment variables are required?

`STRAPI_URL` (default `http://localhost:1337`), `STRAPI_ADMIN_EMAIL`, and `STRAPI_ADMIN_PASSWORD` are recommended. `STRAPI_API_TOKEN` is optional. Set `STRAPI_DEV_MODE=true` to enable development mode features.

3

What are the file size limits for media uploads?

The `upload_media` tool has a ~1MB base64 size limit (~750KB file) to prevent context overflow. `upload_media_from_path` supports files up to 10MB from local file paths.

4

How can I filter, paginate, and sort entries?

The `get_entries` tool supports `filters` (e.g., `{"title": {"$contains": "hello"}}`), `pagination` (`page`, `pageSize`), `sort` (e.g., `["title:asc"]`), and `populate` for relations.

5

What is the purpose of STRAPI_DEV_MODE?

Setting `STRAPI_DEV_MODE` to `true` enables development mode features in Strapi, allowing the server to access additional functionality during local development. It defaults to `false`.

コメント

「その他」の他のコンテンツ

YOUB

Connect your AI assistant to your own YOUB endurance training account. Read recorded activities with full analysis, planned sessions, race goals, wellness (HRV, sleep, resting HR) and your athletic profile. Create or mov

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

Blender

ahujasid

Open-source MCP to use Blender with any LLM

23.6K

Glasswarp

Glasswarp

See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.

Hindsight

vectorize-io

Hindsight: Agent Memory That Learns

50.4K
広告掲載