Skip to main content
Glama

Strapi MCP 服务器

用于与 Strapi CMS 交互的模型上下文协议服务器。该服务器使 AI 助手能够通过标准化接口与您的 Strapi 实例进行交互,支持内容类型和 REST API 操作。

⚠️重要免责声明:本软件是在人工智能技术的帮助下开发的。它按原样提供,未经彻底测试和验证,请勿在生产环境中使用。代码可能包含错误、安全漏洞或意外行为。仅供研究、学习或开发用途使用,风险自负。

变更日志

版本 2.3.0 - 文档和配置增强

  • 📚 在 CLAUDE.md 中添加了全面的项目文档

  • ⚙️ 扩展配置选项,提供更好的版本检测

  • 🛠️ 增强常见问题的故障排除指南

  • 🔄 详细的 REST API 文档和实际示例

  • 📝 内容管理最佳实践指南

  • 🐛 修复了不同格式模式的版本解析

  • 🔍 使用特定于版本的指导改进错误消息

版本 2.2.0 - 安全和版本处理更新

  • 🔒 添加了严格的写保护策略

  • 🔄 增强版本格式支持(5.*、4.1.5、v4 等)

  • 📚 将文档集成到服务器功能中

  • 🚫 删除了连接提示(现在在功能中)

  • ⚡ 改进的错误处理和验证

  • 🔍 添加了特定版本差异指南

  • 📋 增强服务器功能文档

版本 2.1.0

  • 改进了与 Strapi v4 和 v5 的兼容性

  • 删除自动验证以支持版本之间的不同数据结构

  • 使用特定于版本的提示增强错误消息

  • 简化请求处理,为客户提供更多控制权

  • 更新了两个版本的文档,其中包含清晰的示例

Related MCP server: Directus MCP Server

特征

  • 🔍 Schema 自省

  • 🔄 REST API 支持验证

  • 📸 媒体上传处理

  • 🔐 JWT 身份验证

  • 📝 内容类型管理

  • 🖼️ 图像处理与格式转换

  • 🌐 多服务器支持

  • ✅ 自动模式验证

  • 🔒 写保护策略

  • 📚 集成文档

  • 🔄 版本兼容性管理

安装

您可以在 Claude Desktop 配置中直接将此服务器与 npx 一起使用:

{
  "mcpServers": {
    "strapi": {
      "command": "npx",
      "args": ["-y", "@bschauer/strapi-mcp-server@2.5.0"]
    }
  }
}

配置

在~/.mcp/strapi-mcp-server.config.json创建配置文件:

{
  "myserver": {
    "api_url": "http://localhost:1337",
    "api_key": "your-jwt-token-from-strapi-admin",
    "version": "5.*" // Optional: Specify Strapi version (e.g., "5.*", "4.1.5", "v4")
  }
}

您可以通过将多个 Strapi 实例添加到此文件来配置它们。

版本配置

服务器现在支持多种版本格式:

  • 通配符:“5. ”、“4. ”

  • 具体:“4.1.5”,“5.0.0”

  • 简单:“v4”,“v5”

这有助于服务器提供特定版本的指导并适当地处理 API 差异。

获取 JWT 令牌

  1. 登录您的 Strapi 管理面板

  2. 创建具有适当权限的 API 令牌

  3. 将令牌添加到配置文件中相应的服务器名称下

用法

列出可用服务器

strapi_list_servers();
// Now includes version information and differences between v4 and v5

内容类型

// Get all content types from a specific server
strapi_get_content_types({
  server: "myserver",
});

// Get components with pagination
strapi_get_components({
  server: "myserver",
  page: 1,
  pageSize: 25,
});

REST API

REST API 提供了全面的 CRUD 操作,具有内置验证和特定于版本的处理:

// Query content with filters
strapi_rest({
  server: "myserver",
  endpoint: "api/articles",
  method: "GET",
  params: {
    filters: {
      title: {
        $contains: "search term",
      },
    },
  },
});

// Create new content
strapi_rest({
  server: "myserver",
  endpoint: "api/articles",
  method: "POST",
  body: {
    data: {
      title: "New Article",
      content: "Article content",
      category: "news",
    },
  },
});

// Update content
strapi_rest({
  server: "myserver",
  endpoint: "api/articles/123",
  method: "PUT",
  body: {
    data: {
      title: "Updated Title",
      content: "Updated content",
    },
  },
});

// Delete content
strapi_rest({
  server: "myserver",
  endpoint: "api/articles/123",
  method: "DELETE",
});

媒体上传

// Upload image with automatic optimization
strapi_upload_media({
  server: "myserver",
  url: "https://example.com/image.jpg",
  format: "webp",
  quality: 80,
  metadata: {
    name: "My Image",
    caption: "Image Caption",
    alternativeText: "Alt Text",
  },
});

版本差异(v4 与 v5)

服务器自动处理的 Strapi 版本之间的主要区别:

v4

  • 使用数字 ID

  • 嵌套属性结构

  • 响应中的数据包装器

  • 传统的 REST 模式

  • 外部 i18n 插件

v5

  • 基于文档的身份证件

  • 扁平数据结构

  • 直接属性访问

  • 增强的 JWT 安全性

  • 集成 i18n 支持

  • 新的文档服务 API

安全功能

写保护策略

服务器实施严格的写保护策略:

  • 所有写入操作都需要明确授权

  • 受保护的操作包括:

    • POST(创建)

    • PUT(更新)

    • 删除

    • 媒体上传

  • 每个操作都经过记录和验证

最佳实践

  1. 始终先使用strapi_get_content_types检查架构

  2. 对端点使用适当的复数/单数形式

  3. 在查询中包含错误处理

  4. 上传前验证 URL

  5. 从最少的查询开始,仅在需要时添加人口

  6. 更新时始终包含完整的数据对象

  7. 使用过滤器优化查询性能

  8. 利用内置架构验证

  9. 检查操作的版本兼容性

  10. 遵循写保护策略指南

REST API 技巧

过滤

// Filter by field value
params: {
  filters: {
    title: "Exact Match";
  }
}

// Contains filter
params: {
  filters: {
    title: {
      $contains: "partial";
    }
  }
}

// Multiple conditions
params: {
  filters: {
    $and: [{ category: "news" }, { published: true }];
  }
}

排序

params: {
  sort: ["createdAt:desc"];
}

分页

params: {
  pagination: {
    page: 1,
    pageSize: 25
  }
}

人口

// Basic request without population
params: {
}

// Selective population when needed
params: {
  populate: ["category"];
}

// Detailed population with field selection
params: {
  populate: {
    category: {
      fields: ["name", "slug"];
    }
  }
}

故障排除

常见问题及解决方案:

  1. 404 错误

    • 检查端点复数/单数形式

    • 验证内容类型是否存在

    • 确保 API URL 正确

    • 检查是否使用正确的 ID 格式(数字或基于文档)

  2. 身份验证问题

    • 验证 JWT 令牌是否有效

    • 检查令牌权限

    • 确保令牌未过期

  3. 版本相关问题

    • 验证配置中的版本规范

    • 检查数据结构是否匹配版本

    • 查看版本差异文档

  4. 写保护错误

    • 确保操作已获得授权

    • 检查操作是否受到保护

    • 验证请求是否符合安全策略

贡献

欢迎贡献代码!欢迎提交 Pull 请求。

执照

麻省理工学院

Available Tools

5 tools
strapi_get_componentsA

Get all components from Strapi with pagination support. Returns both component data and pagination metadata (page, pageSize, total, pageCount).

Inputs: server (required), page, pageSize.

strapi_get_content_typesB

Get all content types from Strapi. Returns the complete schema of all content types.

Initialization Steps (ALWAYS DO FIRST)

  1. Get schema and analyze with this tool

  2. Capture Content Types and structures

  3. Remember endpoint names (pluralName/singularName)

  4. Document fields and types

  5. Identify relations

  6. Consider required fields and validations

Schema Conventions

  • singularName: Used for single item queries (e.g., "article")

  • pluralName: Used for collection endpoints (e.g., "articles")

  • collectionName: Database collection name

Endpoint Patterns

  • Collection: GET /api/{pluralName}

  • Single: GET /api/{pluralName}/{id}

  • Create: POST /api/{pluralName}

  • Update: PUT /api/{pluralName}/{id}

  • Delete: DELETE /api/{pluralName}/{id}

Inputs: server (required).

strapi_list_serversC

List all available Strapi servers from the configuration.

Security Policy

STRICT_USER_AUTHORIZATION_REQUIRED: No write operations without explicit user authorization. Protected operations: POST (Create), PUT (Update), DELETE (Delete), Media Upload. All write operations require userAuthorized: true parameter.

Strapi Version Support

Supports both Strapi v4 and v5 with automatic version detection.

Version Differences

  • v4: Numeric IDs, nested attributes under 'attributes', data wrapper in responses

  • v5: Document-based IDs (documentId), flat structure, direct attribute access

Common Errors

  • 404: Using numeric ID instead of documentId, wrong plural/singular form

  • 405: Incorrect endpoint (/article instead of /articles)

  • 400: Missing data wrapper in request body

Best Practices

  1. Always check schema first with strapi_get_content_types

  2. Use documentId (not numeric id) for Strapi v5

  3. Always use data wrapper for updates: { data: { field: value } }

  4. Use pluralName for collection endpoints (api/articles)

  5. Validate URLs with webtools before using them

Inputs: no named parameters.

strapi_restA

Execute REST API requests against Strapi endpoints. IMPORTANT: All write operations (POST, PUT, DELETE) require explicit user authorization via the userAuthorized parameter.

Reading Data

params: { populate: ['SEO'] } // Populate a component params: { populate: { SEO: { fields: ['Title', 'seoDescription'] } } } // With field selection params: { filters: { title: { $contains: 'search' } } } // Filter results params: { sort: ['createdAt:desc'] } // Sort results params: { pagination: { page: 1, pageSize: 10 } } // Pagination

Writing Data (REQUIRES userAuthorized: true)

body: { data: { componentName: { Title: 'value' }, // Single component componentName: [{ field: 'value' }] // Repeatable component } }

Debugging Guide

  • 404 Error: Check plural/singular form, use documentId not numeric id

  • 400 Error: Check if data wrapper is present in body

  • 405 Error: Check endpoint format (/articles not /article)

  • URL Errors: Validate URLs with webtools first

  • ID Problems: Use documentId for Strapi v5

Strapi v5 Specifics

  • Use documentId instead of numeric id

  • Direct attribute access (no nested attributes)

  • No data wrapper in GET responses

Inputs: server (required), endpoint (required), body, method, +2 more.

strapi_upload_mediaA

Upload media to Strapi's media library from a URL with format conversion, quality control, and metadata options. IMPORTANT: This is a write operation that REQUIRES explicit user authorization via the userAuthorized parameter.

Upload Steps

  1. Upload via strapi_upload_media with URL and metadata

  2. Get image ID from response

  3. Link to content using strapi_rest PUT request

Linking Images to Content (Strapi v5)

After upload, use PUT request to link: { "method": "PUT", "endpoint": "api/articles/{documentId}", "body": { "data": { "images": ["imageId"] } }, "userAuthorized": true }

Inputs: url (required), server (required), format, quality, +2 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.7/5.0

    Scored across 5 tools

    Disambiguation5/5

    Each tool targets a distinct function: schema introspection (components vs content types), server listing, generic REST execution, and media upload. No overlap in purposes.

    Naming Consistency4/5

    Tools follow a strapi_verb_noun pattern mostly (get_components, get_content_types, list_servers, upload_media), but 'strapi_rest' deviates by using an acronym instead of a verb. Overall consistent.

    Tool Count5/5

    5 tools is well-scoped for a Strapi MCP server, covering schema discovery, generic API access, media upload, and server management without superfluous tools.

    Completeness4/5

    The tool set provides schema introspection and a generic REST tool for all CRUD operations, plus media upload. Minor gap: no dedicated tool for content entry listing, but the REST tool covers it with schema guidance.

    Maintenance

    ActivityStale
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    Appeared in Searches