@deventerprisesoftware/scrapi-mcp 0.0.3 → 0.1.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (3) hide show
  1. package/README.md +20 -4
  2. package/dist/index.js +155 -60
  3. package/package.json +46 -42
package/README.md CHANGED
@@ -1,7 +1,13 @@
1
+ ![ScrAPI logo](https://raw.githubusercontent.com/DevEnterpriseSoftware/scrapi-sdk-dotnet/master/icon_small.png)
1
2
 
2
3
  # ScrAPI MCP Server
3
4
 
4
- MCP server for using ScrAPI to scrape web pages.
5
+ [![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](https://opensource.org/licenses/MIT)
6
+ [![NPM Downloads](https://img.shields.io/npm/dm/@deventerprisesoftware/scrapi-mcp)](https://www.npmjs.com/package/@deventerprisesoftware/scrapi-mcp)
7
+ [![Docker Pulls](https://img.shields.io/docker/pulls/deventerprisesoftware/scrapi-mcp)](https://hub.docker.com/r/deventerprisesoftware/scrapi-mcp)
8
+ [![smithery badge](https://smithery.ai/badge/@DevEnterpriseSoftware/scrapi-mcp)](https://smithery.ai/server/@DevEnterpriseSoftware/scrapi-mcp)
9
+
10
+ MCP server for using [ScrAPI](https://scrapi.tech) to scrape web pages.
5
11
 
6
12
  ScrAPI is your ultimate web scraping solution, offering powerful, reliable, and easy-to-use features to extract data from any website effortlessly.
7
13
 
@@ -29,6 +35,14 @@ Optionally get an API key from the [ScrAPI website](https://scrapi.tech).
29
35
 
30
36
  Without an API key you will be limited to one concurrent call and twenty free calls per day with minimal queuing capabilities.
31
37
 
38
+ ### Cloud Server
39
+
40
+ The ScrAPI MCP Server is also available in the cloud over SSE at https://api.scrapi.tech/mcp/sse and streamable HTTP at https://api.scrapi.tech/mcp
41
+
42
+ Cloud MCP servers are not widely supported yet but you can access this directly from your own custom clients or use [MCP Inspector](https://github.com/modelcontextprotocol/inspector) to test it. There is currently no facility to pass through your API key when connecting to the cloud MCP server.
43
+
44
+ ![MCP-Inspector](https://raw.githubusercontent.com/DevEnterpriseSoftware/scrapi-mcp/master/img/mcp-inspector.jpg)
45
+
32
46
  ### Usage with Claude Desktop
33
47
 
34
48
  Add the following to your `claude_desktop_config.json`:
@@ -38,7 +52,7 @@ Add the following to your `claude_desktop_config.json`:
38
52
  ```json
39
53
  {
40
54
  "mcpServers": {
41
- "scrapi": {
55
+ "ScrAPI": {
42
56
  "command": "docker",
43
57
  "args": [
44
58
  "run",
@@ -56,12 +70,12 @@ Add the following to your `claude_desktop_config.json`:
56
70
  }
57
71
  ```
58
72
 
59
- ### NPX
73
+ #### NPX
60
74
 
61
75
  ```json
62
76
  {
63
77
  "mcpServers": {
64
- "scrapi": {
78
+ "ScrAPI": {
65
79
  "command": "npx",
66
80
  "args": [
67
81
  "-y",
@@ -75,6 +89,8 @@ Add the following to your `claude_desktop_config.json`:
75
89
  }
76
90
  ```
77
91
 
92
+ ![Claude-Desktop](https://raw.githubusercontent.com/DevEnterpriseSoftware/scrapi-mcp/master/img/claude-desktop.jpg)
93
+
78
94
  ## Build
79
95
 
80
96
  Docker build:
package/dist/index.js CHANGED
@@ -1,30 +1,108 @@
1
1
  #!/usr/bin/env node
2
+ import express from "express";
3
+ import cors from "cors";
2
4
  import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
3
5
  import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
6
+ import { StreamableHTTPServerTransport } from "@modelcontextprotocol/sdk/server/streamableHttp.js";
4
7
  import { z } from "zod";
8
+ const PORT = process.env.PORT || 5000;
5
9
  const SCRAPI_API_KEY = process.env.SCRAPI_API_KEY || "00000000-0000-0000-0000-000000000000";
6
10
  const SCRAPI_SERVER_NAME = "ScrAPI MCP Server";
7
- const SCRAPI_SERVER_VERSION = "0.0.1";
8
- const server = new McpServer({
9
- name: SCRAPI_SERVER_NAME,
10
- version: SCRAPI_SERVER_VERSION,
11
+ const SCRAPI_SERVER_VERSION = "0.1.0";
12
+ const app = express();
13
+ app.use(cors({
14
+ origin: "*",
15
+ exposedHeaders: ["Mcp-Session-Id", "mcp-protocol-version"],
16
+ allowedHeaders: ["Content-Type", "mcp-session-id"],
17
+ }));
18
+ app.use(express.json());
19
+ // Define session configuration schema
20
+ export const configSchema = z.object({
21
+ scrapiApiKey: z.string().optional().describe("ScrAPI API key for scraping. Leave empty for default limited usage."),
11
22
  });
12
- server.tool("scrape_url_html", "Use a URL to scrape a website using the ScrAPI service and retrieve the result as HTML. " +
13
- "Use this for scraping website content that is difficult to access because of bot detection, captchas or even geolocation restrictions. " +
14
- "The result will be in HTML which is preferable if advanced parsing is required.", { url: z.string().url({ message: "Invalid URL" }) }, async ({ url }) => await scrapeUrl(url, "HTML"));
15
- server.tool("scrape_url_markdown", "Use a URL to scrape a website using the ScrAPI service and retrieve the result as Markdown. " +
16
- "Use this for scraping website content that is difficult to access because of bot detection, captchas or even geolocation restrictions. " +
17
- "The result will be in Markdown which is preferable if the text content of the webpage is important and not the structural information of the page.", { url: z.string().url({ message: "Invalid URL" }) }, async ({ url }) => await scrapeUrl(url, "Markdown"));
18
- async function scrapeUrl(url, format) {
19
- var body = {
20
- url: url,
21
- useBrowser: true,
22
- solveCaptchas: true,
23
- acceptDialogs: true,
24
- proxyType: "Residential",
25
- responseFormat: format,
26
- };
27
- try {
23
+ // Parse configuration from query parameters
24
+ function parseConfig(req) {
25
+ const configParam = req.query.config;
26
+ if (configParam) {
27
+ return JSON.parse(Buffer.from(configParam, "base64").toString());
28
+ }
29
+ return {};
30
+ }
31
+ // Create MCP server with your tools
32
+ export default function createServer({ config, }) {
33
+ const server = new McpServer({
34
+ name: SCRAPI_SERVER_NAME,
35
+ version: SCRAPI_SERVER_VERSION,
36
+ });
37
+ server.registerTool("scrape_url_html", {
38
+ title: "Scrape URL and respond with HTML",
39
+ description: "Use a URL to scrape a website using the ScrAPI service and retrieve the result as HTML. " +
40
+ "Use this for scraping website content that is difficult to access because of bot detection, captchas or even geolocation restrictions. " +
41
+ "The result will be in HTML which is preferable if advanced parsing is required.",
42
+ inputSchema: {
43
+ url: z
44
+ .string()
45
+ .url({ message: "Invalid URL" })
46
+ .describe("The URL to scrape"),
47
+ },
48
+ }, async ({ url }) => await scrapeUrl(url, "HTML"));
49
+ server.registerTool("scrape_url_markdown", {
50
+ title: "Scrape URL and respond with Markdown",
51
+ description: "Use a URL to scrape a website using the ScrAPI service and retrieve the result as Markdown. " +
52
+ "Use this for scraping website content that is difficult to access because of bot detection, captchas or even geolocation restrictions. " +
53
+ "The result will be in Markdown which is preferable if the text content of the webpage is important and not the structural information of the page.",
54
+ inputSchema: {
55
+ url: z
56
+ .string()
57
+ .url({ message: "Invalid URL" })
58
+ .describe("The URL to scrape"),
59
+ },
60
+ }, async ({ url }) => await scrapeUrl(url, "Markdown"));
61
+ async function scrapeUrl(url, format) {
62
+ var body = {
63
+ url: url,
64
+ useBrowser: true,
65
+ solveCaptchas: true,
66
+ acceptDialogs: true,
67
+ proxyType: "Residential",
68
+ responseFormat: format,
69
+ };
70
+ try {
71
+ const response = await fetch("https://api.scrapi.tech/v1/scrape", {
72
+ method: "POST",
73
+ headers: {
74
+ "User-Agent": `${SCRAPI_SERVER_NAME} - ${SCRAPI_SERVER_VERSION}`,
75
+ "Content-Type": "application/json",
76
+ "X-API-KEY": config.scrapiApiKey || SCRAPI_API_KEY,
77
+ },
78
+ body: JSON.stringify(body),
79
+ signal: AbortSignal.timeout(30000),
80
+ });
81
+ const data = await response.text();
82
+ if (response.ok) {
83
+ return {
84
+ content: [
85
+ {
86
+ type: "text",
87
+ mimeType: `text/${format.toLowerCase()}`,
88
+ text: data,
89
+ },
90
+ ],
91
+ };
92
+ }
93
+ return {
94
+ content: [
95
+ {
96
+ type: "text",
97
+ text: data,
98
+ },
99
+ ],
100
+ isError: true,
101
+ };
102
+ }
103
+ catch (error) {
104
+ console.error("Error calling API:", error);
105
+ }
28
106
  const response = await fetch("https://api.scrapi.tech/v1/scrape", {
29
107
  method: "POST",
30
108
  headers: {
@@ -36,57 +114,74 @@ async function scrapeUrl(url, format) {
36
114
  signal: AbortSignal.timeout(30000),
37
115
  });
38
116
  const data = await response.text();
39
- if (response.ok) {
40
- return {
41
- content: [
42
- {
43
- type: "text",
44
- mimeType: `text/${format.toLowerCase()}`,
45
- text: data,
46
- },
47
- ],
48
- };
49
- }
50
117
  return {
51
118
  content: [
52
119
  {
53
120
  type: "text",
121
+ mimeType: `text/${format.toLowerCase()}`,
54
122
  text: data,
55
123
  },
56
124
  ],
57
- isError: true,
58
125
  };
59
126
  }
127
+ return server.server;
128
+ }
129
+ app.all("/mcp", async (req, res) => {
130
+ try {
131
+ // Parse configuration
132
+ const rawConfig = parseConfig ? parseConfig(req) : {};
133
+ // Validate and parse configuration
134
+ const config = configSchema
135
+ ? configSchema.parse({ scrapiApiKey: rawConfig.scrapiApiKey || SCRAPI_API_KEY })
136
+ : {};
137
+ const server = createServer({ config });
138
+ const transport = new StreamableHTTPServerTransport({
139
+ sessionIdGenerator: undefined,
140
+ });
141
+ // Clean up on request close
142
+ res.on("close", () => {
143
+ transport.close();
144
+ server.close();
145
+ });
146
+ await server.connect(transport);
147
+ await transport.handleRequest(req, res, req.body);
148
+ }
60
149
  catch (error) {
61
- console.error("Error calling API:", error);
150
+ console.error("Error handling MCP request:", error);
151
+ if (!res.headersSent) {
152
+ res.status(500).json({
153
+ jsonrpc: "2.0",
154
+ error: { code: -32603, message: "Internal server error" },
155
+ id: null,
156
+ });
157
+ }
62
158
  }
63
- const response = await fetch("https://api.scrapi.tech/v1/scrape", {
64
- method: "POST",
65
- headers: {
66
- "User-Agent": `${SCRAPI_SERVER_NAME} - ${SCRAPI_SERVER_VERSION}`,
67
- "Content-Type": "application/json",
68
- "X-API-KEY": SCRAPI_API_KEY,
69
- },
70
- body: JSON.stringify(body),
71
- signal: AbortSignal.timeout(30000),
72
- });
73
- const data = await response.text();
74
- return {
75
- content: [
76
- {
77
- type: "text",
78
- mimeType: `text/${format.toLowerCase()}`,
79
- text: data,
159
+ });
160
+ // Main function to start the server in the appropriate mode
161
+ async function main() {
162
+ const transport = process.env.TRANSPORT || "stdio";
163
+ if (transport === "http") {
164
+ // Run in HTTP mode
165
+ app.listen(PORT, () => {
166
+ console.log(`MCP HTTP Server listening on port ${PORT}`);
167
+ });
168
+ }
169
+ else {
170
+ const scrapiApiKey = SCRAPI_API_KEY;
171
+ // Create server with configuration
172
+ const server = createServer({
173
+ config: {
174
+ scrapiApiKey,
80
175
  },
81
- ],
82
- };
83
- }
84
- try {
85
- const transport = new StdioServerTransport();
86
- await server.connect(transport);
87
- console.error(`${SCRAPI_SERVER_NAME} running on stdio.`);
176
+ });
177
+ // Start receiving messages on stdin and sending messages on stdout
178
+ const stdioTransport = new StdioServerTransport();
179
+ await server.connect(stdioTransport);
180
+ console.error("MCP Server running in stdio mode");
181
+ }
88
182
  }
89
- catch (error) {
90
- console.error("Fatal error running server:", error);
183
+ // Start the server
184
+ main().catch((error) => {
185
+ console.error("Server error:", error);
91
186
  process.exit(1);
92
- }
187
+ });
package/package.json CHANGED
@@ -1,42 +1,46 @@
1
- {
2
- "name": "@deventerprisesoftware/scrapi-mcp",
3
- "version": "0.0.3",
4
- "description": "MCP server for using ScrAPI to scrape web pages.",
5
- "keywords": [
6
- "mcp",
7
- "web scraper",
8
- "web scraping",
9
- "web data extractor",
10
- "claude",
11
- "ai"
12
- ],
13
- "homepage": "https://scrapi.tech",
14
- "bugs": {
15
- "url": "https://github.com/DevEnterpriseSoftware/scrapi-mcp/issues"
16
- },
17
- "repository": {
18
- "type": "git",
19
- "url": "git+https://github.com/DevEnterpriseSoftware/scrapi-mcp.git"
20
- },
21
- "license": "MIT",
22
- "author": "DevEnterprise Software (https://deventerprise.com)",
23
- "type": "module",
24
- "bin": "dist/index.js",
25
- "files": [
26
- "dist"
27
- ],
28
- "scripts": {
29
- "build": "tsc && shx chmod +x dist/*.js",
30
- "prepare": "npm run build",
31
- "watch": "tsc --watch"
32
- },
33
- "dependencies": {
34
- "@modelcontextprotocol/sdk": "^1.9.0",
35
- "zod": "^3.24.2"
36
- },
37
- "devDependencies": {
38
- "@types/node": "^22.14.0",
39
- "shx": "^0.4.0",
40
- "typescript": "^5.8.3"
41
- }
42
- }
1
+ {
2
+ "name": "@deventerprisesoftware/scrapi-mcp",
3
+ "version": "0.1.0",
4
+ "description": "MCP server for using ScrAPI to scrape web pages.",
5
+ "keywords": [
6
+ "mcp",
7
+ "web scraper",
8
+ "web scraping",
9
+ "web data extractor",
10
+ "claude",
11
+ "ai"
12
+ ],
13
+ "homepage": "https://scrapi.tech",
14
+ "bugs": {
15
+ "url": "https://github.com/DevEnterpriseSoftware/scrapi-mcp/issues"
16
+ },
17
+ "repository": {
18
+ "type": "git",
19
+ "url": "git+https://github.com/DevEnterpriseSoftware/scrapi-mcp.git"
20
+ },
21
+ "license": "MIT",
22
+ "author": "DevEnterprise Software (https://deventerprise.com)",
23
+ "type": "module",
24
+ "bin": "dist/index.js",
25
+ "files": [
26
+ "dist"
27
+ ],
28
+ "scripts": {
29
+ "build": "tsc && shx chmod +x dist/*.js",
30
+ "prepare": "npm run build",
31
+ "watch": "tsc --watch"
32
+ },
33
+ "dependencies": {
34
+ "@modelcontextprotocol/sdk": "^1.17.4",
35
+ "cors": "^2.8.5",
36
+ "express": "^5.1.0",
37
+ "zod": "^3.25.76"
38
+ },
39
+ "devDependencies": {
40
+ "@types/cors": "^2.8.19",
41
+ "@types/express": "^5.0.3",
42
+ "@types/node": "^24.3.0",
43
+ "shx": "^0.4.0",
44
+ "typescript": "^5.9.2"
45
+ }
46
+ }