We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/jackdark425/aigroup-mdtoword-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
// 为Cloudflare Workers环境适配的MCP服务器
// 使用 shared McpServer 和 CloudflareWorkerTransport
import { createMcpServer } from './mcp-server.js';
import { CloudflareWorkerTransport } from './utils/cloudflare-transport.js';
import { DocxMarkdownConverterWorker } from './converter/markdown-worker.js';
import { z } from 'zod';
// 定义Cloudflare Workers需要的类型
interface ExecutionContext {
waitUntil(promise: Promise<any>): void;
passThroughOnException(): void;
}
export interface Env {
BUCKET: any;
}
// 定义直接API的输入Schema
const DirectApiSchema = z.object({
markdown: z.string().describe('Markdown格式的文本内容'),
filename: z.string().regex(/\.docx$/).describe('输出的Word文档文件名，必须以.docx结尾'),
template: z.object({
type: z.enum(['preset']),
presetId: z.string()
}).optional().describe('模板配置'),
styleConfig: z.object({
theme: z.object({
name: z.string().optional(),
colors: z.object({
primary: z.string().regex(/^[0-9A-Fa-f]{6}$/).optional(),
secondary: z.string().regex(/^[0-9A-Fa-f]{6}$/).optional(),
text: z.string().regex(/^[0-9A-Fa-f]{6}$/).optional(),
}).optional(),
}).optional(),
watermark: z.object({
text: z.string(),
font: z.string().optional(),
size: z.number().min(1).max(200).optional(),
color: z.string().regex(/^[0-9A-Fa-f]{6}$/).optional(),
opacity: z.number().min(0).max(1).optional(),
rotation: z.number().min(-90).max(90).optional(),
}).optional(),
tableOfContents: z.object({
enabled: z.boolean().optional(),
title: z.string().optional(),
levels: z.array(z.number().min(1).max(6)).optional(),
showPageNumbers: z.boolean().optional(),
tabLeader: z.enum(['dot', 'hyphen', 'underscore', 'none']).optional(),
}).optional(),
headerFooter: z.object({
header: z.object({
content: z.string().optional(),
alignment: z.enum(['left', 'center', 'right', 'both']).optional(),
}).optional(),
footer: z.object({
content: z.string().optional(),
showPageNumber: z.boolean().optional(),
pageNumberFormat: z.string().optional(),
showTotalPages: z.boolean().optional(),
totalPagesFormat: z.string().optional(),
alignment: z.enum(['left', 'center', 'right', 'both']).optional(),
}).optional(),
firstPageHeader: z.object({
content: z.string().optional(),
alignment: z.enum(['left', 'center', 'right', 'both']).optional(),
}).optional(),
firstPageFooter: z.object({
content: z.string().optional(),
showPageNumber: z.boolean().optional(),
pageNumberFormat: z.string().optional(),
showTotalPages: z.boolean().optional(),
totalPagesFormat: z.string().optional(),
alignment: z.enum(['left', 'center', 'right', 'both']).optional(),
}).optional(),
evenPageHeader: z.object({
content: z.string().optional(),
alignment: z.enum(['left', 'center', 'right', 'both']).optional(),
}).optional(),
evenPageFooter: z.object({
content: z.string().optional(),
showPageNumber: z.boolean().optional(),
pageNumberFormat: z.string().optional(),
showTotalPages: z.boolean().optional(),
totalPagesFormat: z.string().optional(),
alignment: z.enum(['left', 'center', 'right', 'both']).optional(),
}).optional(),
differentFirstPage: z.boolean().optional(),
differentOddEven: z.boolean().optional(),
pageNumberStart: z.number().optional(),
pageNumberFormatType: z.enum(['decimal', 'upperRoman', 'lowerRoman', 'upperLetter', 'lowerLetter']).optional(),
}).optional(),
tableStyles: z.object({
default: z.object({
columnWidths: z.array(z.number()).optional(),
cellAlignment: z.object({
horizontal: z.enum(['left', 'center', 'right']).optional(),
vertical: z.enum(['top', 'center', 'bottom']).optional(),
}).optional(),
stripedRows: z.object({
enabled: z.boolean().optional(),
oddRowShading: z.string().regex(/^[0-9A-Fa-f]{6}$/).optional(),
evenRowShading: z.string().regex(/^[0-9A-Fa-f]{6}$/).optional(),
}).optional(),
}).optional(),
}).optional(),
imageStyles: z.object({
default: z.object({
maxWidth: z.number().optional(),
maxHeight: z.number().optional(),
maintainAspectRatio: z.boolean().optional(),
alignment: z.enum(['left', 'center', 'right']).optional(),
border: z.object({
color: z.string().regex(/^[0-9A-Fa-f]{6}$/).optional(),
width: z.number().optional(),
style: z.enum(['single', 'double', 'dotted', 'dashed']).optional(),
}).optional(),
}).optional(),
}).optional(),
document: z.object({
defaultFont: z.string().optional(),
defaultSize: z.number().optional(),
}).optional(),
}).optional().describe('样式配置对象'),
});
export default {
async fetch(request: Request, env: Env, ctx: ExecutionContext): Promise<Response> {
try {
const url = new URL(request.url);
// 健康检查
if (url.pathname === '/health') {
return new Response(JSON.stringify({
status: 'healthy',
service: 'aigroup-mdtoword-mcp',
version: '4.0.2',
timestamp: new Date().toISOString(),
environment: 'cloudflare-worker',
features: ['mcp', 'http-api', 'docx-conversion', 'cloud-storage']
}), {
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'GET, POST, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type'
},
});
}
// 根路径信息
if (url.pathname === '/') {
return new Response(JSON.stringify({
name: 'aigroup-mdtoword-mcp',
version: '4.0.2',
description: 'Markdown to Word conversion service with MCP protocol and HTTP API support (Cloudflare Worker)',
endpoints: {
mcp: '/mcp',
convert: '/convert',
health: '/health',
},
documentation: 'https://github.com/jackdark425/aigroup-mdtoword-mcp',
}), {
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'GET, POST, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type'
},
});
}
// 处理CORS预检请求
if (request.method === 'OPTIONS') {
return new Response(null, {
headers: {
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'GET, POST, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type, mcp-session-id, Authorization',
},
});
}
// 文件下载路由
if (url.pathname.startsWith('/files/')) {
const filename = url.pathname.slice(7); // remove '/files/'
if (!filename) return new Response('Filename missing', { status: 400 });
const object = await env.BUCKET.get(filename);
if (!object) return new Response('File Not Found', { status: 404 });
const headers = new Headers();
object.writeHttpMetadata(headers);
headers.set('etag', object.httpEtag);
return new Response(object.body, { headers });
}
// MCP协议请求
if (url.pathname === '/mcp') {
// 检查请求方法
if (request.method !== 'POST') {
return new Response('Method Not Allowed', {
status: 405,
headers: { 'Allow': 'POST' }
});
}
// 获取请求体
const body = await request.json();
// 创建Transport和Server
// @ts-ignore - body type might rely on exact JSONRPC validation, transport handles it loosely initially
const transport = new CloudflareWorkerTransport(body);
const server = createMcpServer({
name: 'aigroup-mdtoword-mcp',
version: '4.0.2',
ConverterClass: DocxMarkdownConverterWorker,
// Worker environment doesn't support FS, so we don't pass fileSystem handler
cloudStorage: {
async upload(filename: string, content: Uint8Array, contentType: string) {
// Ensure unique filename or overwrite?
// For now, overwrite if same name, or maybe prefix with timestamp to ensure uniqueness?
// Let's use the provided filename but maybe cleanup
await env.BUCKET.put(filename, content, {
httpMetadata: { contentType }
});
// Use the actual host from the request to ensure correct domain in URLs
return `${request.headers.get('X-Forwarded-Proto') || 'https'}://${request.headers.get('Host') || url.host}/files/${filename}`;
}
}
});
// 连接 Transport
await server.connect(transport);
// 等待响应 (由 CloudflareWorkerTransport 捕获)
const response = await transport.response;
return new Response(JSON.stringify(response), {
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'POST, GET, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type, mcp-session-id'
},
});
}
// 直接API端点 - 将请求转发给MCP服务器进行处理
if (url.pathname === '/convert') {
// 检查请求方法
if (request.method !== 'POST') {
return new Response('Method Not Allowed', {
status: 405,
headers: { 'Allow': 'POST' }
});
}
try {
// 解析请求体
const inputData = await request.json();
// 验证输入数据
const parsedData = DirectApiSchema.parse(inputData);
// 构造MCP协议请求
const mcpRequest = {
method: 'tools/call',
jsonrpc: '2.0' as const,
id: 'direct-api-request',
params: {
name: 'markdown_to_docx',
arguments: {
markdown: parsedData.markdown,
filename: parsedData.filename,
template: parsedData.template,
styleConfig: parsedData.styleConfig
}
}
};
// 创建Transport和Server
const transport = new CloudflareWorkerTransport(mcpRequest);
const server = createMcpServer({
name: 'aigroup-mdtoword-mcp',
version: '4.0.2',
ConverterClass: DocxMarkdownConverterWorker,
// Worker environment doesn't support FS, so we don't pass fileSystem handler
cloudStorage: {
async upload(filename: string, content: Uint8Array, contentType: string) {
await env.BUCKET.put(filename, content, {
httpMetadata: { contentType }
});
// Use the actual host from the request to ensure correct domain in URLs
return `${request.headers.get('X-Forwarded-Proto') || 'https'}://${request.headers.get('Host') || url.host}/files/${filename}`;
}
}
});
// 连接 Transport
await server.connect(transport);
// 等待响应 (由 CloudflareWorkerTransport 捕获)
const response = await transport.response;
// 类型守卫来区分错误和成功响应
if ('error' in response) {
return new Response(JSON.stringify({
success: false,
error: response.error.message || 'Unknown error occurred'
}), {
status: 500,
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'POST, GET, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type'
},
});
} else if ('result' in response) {
// 提取结果内容
const result = response.result;
const structuredContent = result?.structuredContent;
// 类型断言以处理structuredContent的类型
if (structuredContent && typeof structuredContent === 'object' && 'success' in structuredContent && structuredContent.success) {
const contentObj = structuredContent as any;
// 返回成功的直接API响应
return new Response(JSON.stringify({
success: true,
filename: contentObj.filename,
downloadUrl: contentObj.url,
size: contentObj.size,
message: contentObj.message
}), {
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'POST, GET, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type'
},
});
} else {
// 错误情况
const errorContent = result?.content?.[0]?.text || 'Unknown error occurred';
return new Response(JSON.stringify({
success: false,
error: errorContent
}), {
status: 500,
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'POST, GET, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type'
},
});
}
} else {
// 响应格式不符合预期
return new Response(JSON.stringify({
success: false,
error: 'Unexpected response format'
}), {
status: 500,
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'POST, GET, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type'
},
});
}
} catch (validationError) {
// 输入验证失败
return new Response(JSON.stringify({
success: false,
error: 'Invalid input data',
details: validationError instanceof Error ? validationError.message : String(validationError)
}), {
status: 400,
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'POST, GET, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type'
},
});
}
}
// OpenAPI规范端点
if (url.pathname === '/openapi.yaml' || url.pathname === '/openapi.json') {
if (url.pathname === '/openapi.yaml') {
// 返回YAML格式的OpenAPI规范
const yamlSpec = `openapi: 3.0.0
info:
title: Markdown转Word转换器
description: 将Markdown格式的文本转换为Word文档(DOCX格式)，支持丰富的样式配置选项。
version: 4.0.2
servers:
- url: ${url.origin}
paths:
/convert:
post:
summary: 转换Markdown为Word文档
description: 将提供的Markdown内容转换为Word文档，并返回下载链接。
operationId: convertMarkdownToWord
requestBody:
required: true
content:
application/json:
schema:
type: object
required:
- markdown
- filename
properties:
markdown:
type: string
description: 要转换的Markdown格式文本内容
example: "# 标题\\n\\n这是一段内容。"
filename:
type: string
description: 输出的Word文档文件名，必须以.docx结尾
pattern: ".*\\\\.docx$"
example: "document.docx"
template:
type: object
description: 模板配置
properties:
type:
type: string
enum: [preset]
description: 模板类型
presetId:
type: string
description: 预设模板ID (如: academic, business, customer-analysis, technical, minimal, enhanced-features)
example:
type: "preset"
presetId: "customer-analysis"
styleConfig:
type: object
description: 样式配置对象
properties:
theme:
type: object
description: 主题配置
properties:
name:
type: string
description: 主题名称
colors:
type: object
properties:
primary:
type: string
pattern: "^[0-9A-Fa-f]{6}$"
description: 主色调（6位十六进制）
secondary:
type: string
pattern: "^[0-9A-Fa-f]{6}$"
description: 辅助色（6位十六进制）
text:
type: string
pattern: "^[0-9A-Fa-f]{6}$"
description: 文本颜色（6位十六进制）
watermark:
type: object
description: 水印配置
properties:
text:
type: string
description: 水印文本
font:
type: string
description: 水印字体
size:
type: number
minimum: 1
maximum: 200
description: 水印字号
color:
type: string
pattern: "^[0-9A-Fa-f]{6}$"
description: 水印颜色（6位十六进制）
opacity:
type: number
minimum: 0
maximum: 1
description: 透明度（0-1）
rotation:
type: number
minimum: -90
maximum: 90
description: 旋转角度（-90到90）
document:
type: object
description: 文档基本配置
properties:
defaultFont:
type: string
description: 默认字体
example: "Arial"
defaultSize:
type: number
description: 默认字号（半点）
example: 24
responses:
'200':
description: 转换成功
content:
application/json:
schema:
type: object
properties:
success:
type: boolean
description: 转换是否成功
filename:
type: string
description: 生成的文件名
downloadUrl:
type: string
description: 文档下载链接
size:
type: number
description: 文件大小（字节）
message:
type: string
description: 操作消息
required:
- success
- filename
- downloadUrl
- size
- message
'400':
description: 请求参数错误
content:
application/json:
schema:
type: object
properties:
success:
type: boolean
description: 操作是否成功
error:
type: string
description: 错误信息
required:
- success
- error
'500':
description: 服务器内部错误
content:
application/json:
schema:
type: object
properties:
success:
type: boolean
description: 操作是否成功
error:
type: string
description: 错误信息
required:
- success
- error
x-openai-isConsequential: false`;
return new Response(yamlSpec, {
headers: {
'Content-Type': 'text/yaml',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'GET, POST, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type'
},
});
} else {
// 返回JSON格式的OpenAPI规范
const jsonSpec = {
openapi: '3.0.0',
info: {
title: 'Markdown转Word转换器',
description: '将Markdown格式的文本转换为Word文档(DOCX格式)，支持丰富的样式配置选项。',
version: '4.0.2'
},
servers: [
{ url: url.origin }
],
paths: {
'/convert': {
post: {
summary: '转换Markdown为Word文档',
description: '将提供的Markdown内容转换为Word文档，并返回下载链接。',
operationId: 'convertMarkdownToWord',
requestBody: {
required: true,
content: {
'application/json': {
schema: {
type: 'object',
required: ['markdown', 'filename'],
properties: {
markdown: {
type: 'string',
description: '要转换的Markdown格式文本内容',
example: '# 标题\\n\\n这是一段内容。'
},
filename: {
type: 'string',
description: '输出的Word文档文件名，必须以.docx结尾',
pattern: '.*\\.docx$',
example: 'document.docx'
},
template: {
type: 'object',
description: '模板配置',
properties: {
type: {
type: 'string',
enum: ['preset'],
description: '模板类型'
},
presetId: {
type: 'string',
description: '预设模板ID (如: academic, business, customer-analysis, technical, minimal, enhanced-features)'
}
},
example: {
type: 'preset',
presetId: 'customer-analysis'
}
},
styleConfig: {
type: 'object',
description: '样式配置对象',
properties: {
theme: {
type: 'object',
description: '主题配置',
properties: {
name: {
type: 'string',
description: '主题名称'
},
colors: {
type: 'object',
properties: {
primary: {
type: 'string',
pattern: '^[0-9A-Fa-f]{6}$',
description: '主色调（6位十六进制）'
},
secondary: {
type: 'string',
pattern: '^[0-9A-Fa-f]{6}$',
description: '辅助色（6位十六进制）'
},
text: {
type: 'string',
pattern: '^[0-9A-Fa-f]{6}$',
description: '文本颜色（6位十六进制）'
}
}
}
}
},
watermark: {
type: 'object',
description: '水印配置',
properties: {
text: {
type: 'string',
description: '水印文本'
},
font: {
type: 'string',
description: '水印字体'
},
size: {
type: 'number',
minimum: 1,
maximum: 200,
description: '水印字号'
},
color: {
type: 'string',
pattern: '^[0-9A-Fa-f]{6}$',
description: '水印颜色（6位十六进制）'
},
opacity: {
type: 'number',
minimum: 0,
maximum: 1,
description: '透明度（0-1）'
},
rotation: {
type: 'number',
minimum: -90,
maximum: 90,
description: '旋转角度（-90到90）'
}
}
},
document: {
type: 'object',
description: '文档基本配置',
properties: {
defaultFont: {
type: 'string',
description: '默认字体',
example: 'Arial'
},
defaultSize: {
type: 'number',
description: '默认字号（半点）',
example: 24
}
}
}
}
}
}
}
}
}
},
responses: {
'200': {
description: '转换成功',
content: {
'application/json': {
schema: {
type: 'object',
properties: {
success: {
type: 'boolean',
description: '转换是否成功'
},
filename: {
type: 'string',
description: '生成的文件名'
},
downloadUrl: {
type: 'string',
description: '文档下载链接'
},
size: {
type: 'number',
description: '文件大小（字节）'
},
message: {
type: 'string',
description: '操作消息'
}
},
required: ['success', 'filename', 'downloadUrl', 'size', 'message']
}
}
}
},
'400': {
description: '请求参数错误',
content: {
'application/json': {
schema: {
type: 'object',
properties: {
success: {
type: 'boolean',
description: '操作是否成功'
},
error: {
type: 'string',
description: '错误信息'
}
},
required: ['success', 'error']
}
}
}
},
'500': {
description: '服务器内部错误',
content: {
'application/json': {
schema: {
type: 'object',
properties: {
success: {
type: 'boolean',
description: '操作是否成功'
},
error: {
type: 'string',
description: '错误信息'
}
},
required: ['success', 'error']
}
}
}
}
},
'x-openai-isConsequential': false
}
}
}
};
return new Response(JSON.stringify(jsonSpec), {
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'GET, POST, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type'
},
});
}
}
// OpenAI插件清单端点
if (url.pathname === '/.well-known/ai-plugin.json') {
const pluginManifest = {
schema_version: 'v1',
name_for_model: 'markdown_to_word_converter',
name_for_human: 'Markdown转Word转换器',
description_for_model: '将Markdown格式的文本转换为Word文档(DOCX格式)，支持丰富的样式配置选项。',
description_for_human: '将Markdown文档转换为专业的Word文档。',
auth: {
type: 'none'
},
api: {
type: 'openapi',
url: `${url.origin}/openapi.yaml`,
has_user_authentication: false
},
logo_url: `${url.origin}/logo.png`,
contact_email: 'support@example.com',
legal_info_url: 'https://example.com/legal'
};
return new Response(JSON.stringify(pluginManifest), {
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Methods': 'GET, POST, OPTIONS',
'Access-Control-Allow-Headers': 'Content-Type'
},
});
}
// Logo端点 - 返回一个简单的SVG logo
if (url.pathname === '/logo.png') {
// 实际上返回一个SVG，但设置为PNG MIME类型以简化
const svgLogo = `<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 100 100">
<rect width="100" height="100" fill="#3498db"/>
<text x="50%" y="50%" dominant-baseline="middle" text-anchor="middle" font-family="Arial" font-size="16" fill="white">MD→DOCX</text>
</svg>`;
return new Response(svgLogo, {
headers: {
'Content-Type': 'image/svg+xml',
'Access-Control-Allow-Origin': '*',
'Cache-Control': 'public, max-age=3600'
},
});
}
// 未知路径
return new Response('Not Found', {
status: 404,
headers: { 'Content-Type': 'text/plain' }
});
} catch (error) {
console.error('Worker错误:', error);
return new Response(JSON.stringify({
error: 'Internal Server Error',
message: error instanceof Error ? error.message : String(error),
}), {
status: 500,
headers: {
'Content-Type': 'application/json',
'Access-Control-Allow-Origin': '*'
},
});
}
},
};