Upload folder using huggingface_hub

Crawler/Clients/Client_request.py

@@ -0,0 +1,82 @@
+import requests
+import re
+import time
+import os
+import json
+
+RETRY = 3
+SLEEP_BETWEEN_RETRY = 2
+
+def extract_hrefs(response_text):
+    # Extract all hrefs
+    return re.findall(r'"href":"(/client/[^"]+)"', response_text)
+
+def main():
+    output_file = "mcpso_all_hrefs.json"
+    visited = set()
+    href_list = []
+    # Resume: load already saved hrefs
+    if os.path.exists(output_file):
+        with open(output_file, "r", encoding="utf-8") as f:
+            try:
+                href_list = json.load(f)
+                for item in href_list:
+                    visited.add(item["href"])
+            except Exception:
+                pass
+
+    headers = {
+        "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36",
+        "Accept": "*/*",
+        "Accept-Encoding": "gzip, deflate, br, zstd",
+        "Accept-Language": "zh-CN,zh;q=0.9",
+        "Cookie": "Your Cookie",
+        "Next-Url": "/en/servers",
+        "Priority": "u=1, i",
+        "Referer": "https://mcp.so/servers",
+        "Rsc": "1",
+        "Sec-Ch-Ua": '"Chromium";v="136", "Google Chrome";v="136", "Not.A/Brand";v="99"',
+        "Sec-Ch-Ua-Arch": "arm",
+        "Sec-Ch-Ua-Bitness": "64",
+        "Sec-Ch-Ua-Full-Version": "136.0.7103.114",
+        "Sec-Ch-Ua-Full-Version-List": '"Chromium";v="136.0.7103.114", "Google Chrome";v="136.0.7103.114", "Not.A/Brand";v="99.0.0.0"',
+        "Sec-Ch-Ua-Mobile": "?0",
+        "Sec-Ch-Ua-Model": '""',
+        "Sec-Ch-Ua-Platform": '"macOS"',
+        "Sec-Ch-Ua-Platform-Version": '"15.3.0"',
+        "Sec-Fetch-Dest": "empty",
+        "Sec-Fetch-Mode": "cors",
+        "Sec-Fetch-Site": "same-origin"
+    }
+
+    for page in range(1, 477):
+        url = f"https://mcp.so/clients?page={page}"
+        print(f"Requesting page {page}: {url}")
+        for attempt in range(1, RETRY + 1):
+            try:
+                resp = requests.get(url, headers=headers, timeout=30)
+                if resp.status_code != 200:
+                    print(f"Page {page} failed: HTTP {resp.status_code}, attempt {attempt}/{RETRY}")
+                    time.sleep(SLEEP_BETWEEN_RETRY)
+                    continue
+                hrefs = extract_hrefs(resp.text)
+                new_hrefs = [h for h in hrefs if h not in visited]
+                for h in new_hrefs:
+                    href_list.append({"href": h})
+                    visited.add(h)
+                # Save in real time
+                with open(output_file, "w", encoding="utf-8") as f:
+                    json.dump(href_list, f, ensure_ascii=False, indent=2)
+                print(f"Page {page} got {len(new_hrefs)} new, total {len(href_list)}")
+                time.sleep(1)
+                break
+            except Exception as e:
+                print(f"Page {page} exception: {e}, attempt {attempt}/{RETRY}")
+                time.sleep(SLEEP_BETWEEN_RETRY)
+                continue
+        else:
+            print(f"Page {page} failed after {RETRY} retries.")
+    print(f"All done. Total unique hrefs: {len(href_list)}. Saved to {output_file}")
+
+if __name__ == "__main__":
+    main() 

Crawler/Clients/mcpso_client_detail_request.py

@@ -0,0 +1,142 @@
+import requests
+import re
+import json
+import time
+import os
+
+def extract_current_project(text):
+    # 1. Locate currentProject start position
+    key = '"currentProject":'
+    start = text.find(key)
+    if start == -1:
+        print("currentProject not found!")
+        return None
+    start = start + len(key)
+    # 2. Starting from the first {, use bracket counting method to find matching }
+    while start < len(text) and text[start] not in '{':
+        start += 1
+    if start == len(text):
+        print("currentProject JSON start not found!")
+        return None
+    brace_count = 0
+    end = start
+    for i, c in enumerate(text[start:]):
+        if c == '{':
+            brace_count += 1
+        elif c == '}':
+            brace_count -= 1
+        if brace_count == 0:
+            end = start + i + 1
+            break
+    json_str = text[start:end]
+    try:
+        profile = json.loads(json_str)
+        return profile
+    except Exception as e:
+        print(f"JSON decode error: {e}")
+        return None
+
+def request_server_detail(url, headers):
+    try:
+        resp = requests.get(url, headers=headers, timeout=30)
+        print(f"Status code: {resp.status_code} for {url}")
+        if resp.status_code == 200:
+            profile = extract_current_project(resp.text)
+            return profile
+        else:
+            print(f"Failed to get detail: HTTP {resp.status_code}")
+            return None
+    except Exception as e:
+        print(f"Exception: {e}")
+        return None
+
+def batch_request_servers():
+    # Read mcpso_clients.json
+    servers_path = os.path.join(os.path.dirname(__file__), 'mcpso_clients.json')
+    with open(servers_path, 'r', encoding='utf-8') as f:
+        servers = json.load(f)
+    headers = {
+        "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36",
+        "Accept": "*/*",
+        "Accept-Encoding": "gzip, deflate, br, zstd",
+        "Accept-Language": "zh-CN,zh;q=0.9",
+        "Cookie": "Your Cookie",
+        "Next-Url": "/en/servers",
+        "Priority": "u=1, i",
+        "Referer": "https://mcp.so/servers",
+        "Rsc": "1",
+        "Sec-Ch-Ua": '"Chromium";v="136", "Google Chrome";v="136", "Not.A/Brand";v="99"',
+        "Sec-Ch-Ua-Arch": "arm",
+        "Sec-Ch-Ua-Bitness": "64",
+        "Sec-Ch-Ua-Full-Version": "136.0.7103.114",
+        "Sec-Ch-Ua-Full-Version-List": '"Chromium";v="136.0.7103.114", "Google Chrome";v="136.0.7103.114", "Not.A/Brand";v="99.0.0.0"',
+        "Sec-Ch-Ua-Mobile": "?0",
+        "Sec-Ch-Ua-Model": '""',
+        "Sec-Ch-Ua-Platform": '"macOS"',
+        "Sec-Ch-Ua-Platform-Version": '"15.3.0"',
+        "Sec-Fetch-Dest": "empty",
+        "Sec-Fetch-Mode": "cors",
+        "Sec-Fetch-Site": "same-origin"
+    }
+    for idx, item in enumerate(servers):
+        # Skip already collected items (already have name field)
+        if 'name' in item and 'metadata' in item:
+            continue
+        href = item.get('href')
+        if not href:
+            continue
+        detail_url = f"https://mcp.so{href}"
+        print(f"Requesting: {detail_url}")
+        profile = request_server_detail(detail_url, headers)
+        if not profile:
+            print(f"Skip {href} due to extraction failure.")
+            continue
+        name = profile.get('name')
+        url = profile.get('url')
+        metadata = profile.copy()
+        metadata.pop('name', None)
+        metadata.pop('url', None)
+        item['name'] = name
+        item['url'] = url
+        item['metadata'] = metadata
+        # Write back in real time
+        with open(servers_path, 'w', encoding='utf-8') as f:
+            json.dump(servers, f, ensure_ascii=False, indent=2)
+        print(f"Updated {idx+1}/{len(servers)}: {name}")
+        time.sleep(1)
+    print(f"All servers updated in {servers_path}")
+
+if __name__ == "__main__":
+    # # Single collection (original logic)
+    # url = "https://mcp.so/server/zhipu-web-search/BigModel?_rsc=n713a"
+    # headers = {
+    #     "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36",
+    #     "Accept": "*/*",
+    #     "Accept-Encoding": "gzip, deflate, br, zstd",
+    #     "Accept-Language": "zh-CN,zh;q=0.9",
+    #     "Next-Url": "/en/server/zhipu-web-search/BigModel",
+    #     "Priority": "u=1, i",
+    #     "Referer": "https://mcp.so/server/zhipu-web-search/BigModel",
+    #     "Rsc": "1",
+    #     "Sec-Ch-Ua": '"Chromium";v="136", "Google Chrome";v="136", "Not.A/Brand";v="99"',
+    #     "Sec-Ch-Ua-Arch": "arm",
+    #     "Sec-Ch-Ua-Bitness": "64",
+    #     "Sec-Ch-Ua-Full-Version": "136.0.7103.114",
+    #     "Sec-Ch-Ua-Full-Version-List": '"Chromium";v="136.0.7103.114", "Google Chrome";v="136.0.7103.114", "Not.A/Brand";v="99.0.0.0"',
+    #     "Sec-Ch-Ua-Mobile": "?0",
+    #     "Sec-Ch-Ua-Model": '""',
+    #     "Sec-Ch-Ua-Platform": '"macOS"',
+    #     "Sec-Ch-Ua-Platform-Version": '"15.3.0"',
+    #     "Sec-Fetch-Dest": "empty",
+    #     "Sec-Fetch-Mode": "cors",
+    #     "Sec-Fetch-Site": "same-origin"
+    # }
+    # profile = request_server_detail(url, headers)
+    # if profile:
+    #     with open("server_zhipu-web-search_BigModel_profile.json", "w", encoding="utf-8") as f:
+    #         json.dump(profile, f, ensure_ascii=False, indent=2)
+    #     print("Profile saved to server_zhipu-web-search_BigModel_profile.json")
+    # else:
+    #     print("Profile extraction failed!")
+    # Batch collection
+    batch_request_servers() 

Crawler/Servers/Server_request.py

@@ -0,0 +1,82 @@
+import requests
+import re
+import time
+import os
+import json
+
+RETRY = 3
+SLEEP_BETWEEN_RETRY = 2
+
+def extract_hrefs(response_text):
+    # Extract all hrefs
+    return re.findall(r'"href":"(/server/[^"]+)"', response_text)
+
+def main():
+    output_file = "mcpso_all_hrefs.json"
+    visited = set()
+    href_list = []
+    # Resume: load already saved hrefs
+    if os.path.exists(output_file):
+        with open(output_file, "r", encoding="utf-8") as f:
+            try:
+                href_list = json.load(f)
+                for item in href_list:
+                    visited.add(item["href"])
+            except Exception:
+                pass
+
+    headers = {
+        "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36",
+        "Accept": "*/*",
+        "Accept-Encoding": "gzip, deflate, br, zstd",
+        "Accept-Language": "zh-CN,zh;q=0.9",
+        "Cookie": "Your Cookie",
+        "Next-Url": "/en/servers",
+        "Priority": "u=1, i",
+        "Referer": "https://mcp.so/servers",
+        "Rsc": "1",
+        "Sec-Ch-Ua": '"Chromium";v="136", "Google Chrome";v="136", "Not.A/Brand";v="99"',
+        "Sec-Ch-Ua-Arch": "arm",
+        "Sec-Ch-Ua-Bitness": "64",
+        "Sec-Ch-Ua-Full-Version": "136.0.7103.114",
+        "Sec-Ch-Ua-Full-Version-List": '"Chromium";v="136.0.7103.114", "Google Chrome";v="136.0.7103.114", "Not.A/Brand";v="99.0.0.0"',
+        "Sec-Ch-Ua-Mobile": "?0",
+        "Sec-Ch-Ua-Model": '""',
+        "Sec-Ch-Ua-Platform": '"macOS"',
+        "Sec-Ch-Ua-Platform-Version": '"15.3.0"',
+        "Sec-Fetch-Dest": "empty",
+        "Sec-Fetch-Mode": "cors",
+        "Sec-Fetch-Site": "same-origin"
+    }
+
+    for page in range(1, 477):
+        url = f"https://mcp.so/servers?page={page}"
+        print(f"Requesting page {page}: {url}")
+        for attempt in range(1, RETRY + 1):
+            try:
+                resp = requests.get(url, headers=headers, timeout=30)
+                if resp.status_code != 200:
+                    print(f"Page {page} failed: HTTP {resp.status_code}, attempt {attempt}/{RETRY}")
+                    time.sleep(SLEEP_BETWEEN_RETRY)
+                    continue
+                hrefs = extract_hrefs(resp.text)
+                new_hrefs = [h for h in hrefs if h not in visited]
+                for h in new_hrefs:
+                    href_list.append({"href": h})
+                    visited.add(h)
+                # Save in real time
+                with open(output_file, "w", encoding="utf-8") as f:
+                    json.dump(href_list, f, ensure_ascii=False, indent=2)
+                print(f"Page {page} got {len(new_hrefs)} new, total {len(href_list)}")
+                time.sleep(1)
+                break
+            except Exception as e:
+                print(f"Page {page} exception: {e}, attempt {attempt}/{RETRY}")
+                time.sleep(SLEEP_BETWEEN_RETRY)
+                continue
+        else:
+            print(f"Page {page} failed after {RETRY} retries.")
+    print(f"All done. Total unique hrefs: {len(href_list)}. Saved to {output_file}")
+
+if __name__ == "__main__":
+    main() 

Crawler/Servers/mcpso_server_detail_request.py

@@ -0,0 +1,142 @@
+import requests
+import re
+import json
+import time
+import os
+
+def extract_current_project(text):
+    # 1. Locate currentProject start position
+    key = '"currentProject":'
+    start = text.find(key)
+    if start == -1:
+        print("currentProject not found!")
+        return None
+    start = start + len(key)
+    # 2. Starting from the first {, use bracket counting method to find matching }
+    while start < len(text) and text[start] not in '{':
+        start += 1
+    if start == len(text):
+        print("currentProject JSON start not found!")
+        return None
+    brace_count = 0
+    end = start
+    for i, c in enumerate(text[start:]):
+        if c == '{':
+            brace_count += 1
+        elif c == '}':
+            brace_count -= 1
+        if brace_count == 0:
+            end = start + i + 1
+            break
+    json_str = text[start:end]
+    try:
+        profile = json.loads(json_str)
+        return profile
+    except Exception as e:
+        print(f"JSON decode error: {e}")
+        return None
+
+def request_server_detail(url, headers):
+    try:
+        resp = requests.get(url, headers=headers, timeout=30)
+        print(f"Status code: {resp.status_code} for {url}")
+        if resp.status_code == 200:
+            profile = extract_current_project(resp.text)
+            return profile
+        else:
+            print(f"Failed to get detail: HTTP {resp.status_code}")
+            return None
+    except Exception as e:
+        print(f"Exception: {e}")
+        return None
+
+def batch_request_servers():
+    # Read mcpso_servers.json
+    servers_path = os.path.join(os.path.dirname(__file__), 'mcpso_servers.json')
+    with open(servers_path, 'r', encoding='utf-8') as f:
+        servers = json.load(f)
+    headers = {
+        "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36",
+        "Accept": "*/*",
+        "Accept-Encoding": "gzip, deflate, br, zstd",
+        "Accept-Language": "zh-CN,zh;q=0.9",
+        "Cookie": "Your Cookie",
+        "Next-Url": "/en/server/zhipu-web-search/BigModel",
+        "Priority": "u=1, i",
+        "Referer": "https://mcp.so/server/zhipu-web-search/BigModel",
+        "Rsc": "1",
+        "Sec-Ch-Ua": '"Chromium";v="136", "Google Chrome";v="136", "Not.A/Brand";v="99"',
+        "Sec-Ch-Ua-Arch": "arm",
+        "Sec-Ch-Ua-Bitness": "64",
+        "Sec-Ch-Ua-Full-Version": "136.0.7103.114",
+        "Sec-Ch-Ua-Full-Version-List": '"Chromium";v="136.0.7103.114", "Google Chrome";v="136.0.7103.114", "Not.A/Brand";v="99.0.0.0"',
+        "Sec-Ch-Ua-Mobile": "?0",
+        "Sec-Ch-Ua-Model": '""',
+        "Sec-Ch-Ua-Platform": '"macOS"',
+        "Sec-Ch-Ua-Platform-Version": '"15.3.0"',
+        "Sec-Fetch-Dest": "empty",
+        "Sec-Fetch-Mode": "cors",
+        "Sec-Fetch-Site": "same-origin"
+    }
+    for idx, item in enumerate(servers):
+        # Skip already collected items (already have name field)
+        if 'name' in item and 'metadata' in item:
+            continue
+        href = item.get('href')
+        if not href:
+            continue
+        detail_url = f"https://mcp.so{href}"
+        print(f"Requesting: {detail_url}")
+        profile = request_server_detail(detail_url, headers)
+        if not profile:
+            print(f"Skip {href} due to extraction failure.")
+            continue
+        name = profile.get('name')
+        url = profile.get('url')
+        metadata = profile.copy()
+        metadata.pop('name', None)
+        metadata.pop('url', None)
+        item['name'] = name
+        item['url'] = url
+        item['metadata'] = metadata
+        # Write back in real time
+        with open(servers_path, 'w', encoding='utf-8') as f:
+            json.dump(servers, f, ensure_ascii=False, indent=2)
+        print(f"Updated {idx+1}/{len(servers)}: {name}")
+        time.sleep(1)
+    print(f"All servers updated in {servers_path}")
+
+if __name__ == "__main__":
+    # 
+    # url = "https://mcp.so/server/zhipu-web-search/BigModel?_rsc=n713a"
+    # headers = {
+    #     "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36",
+    #     "Accept": "*/*",
+    #     "Accept-Encoding": "gzip, deflate, br, zstd",
+    #     "Accept-Language": "zh-CN,zh;q=0.9",
+    #     "Next-Url": "/en/server/zhipu-web-search/BigModel",
+    #     "Priority": "u=1, i",
+    #     "Referer": "https://mcp.so/server/zhipu-web-search/BigModel",
+    #     "Rsc": "1",
+    #     "Sec-Ch-Ua": '"Chromium";v="136", "Google Chrome";v="136", "Not.A/Brand";v="99"',
+    #     "Sec-Ch-Ua-Arch": "arm",
+    #     "Sec-Ch-Ua-Bitness": "64",
+    #     "Sec-Ch-Ua-Full-Version": "136.0.7103.114",
+    #     "Sec-Ch-Ua-Full-Version-List": '"Chromium";v="136.0.7103.114", "Google Chrome";v="136.0.7103.114", "Not.A/Brand";v="99.0.0.0"',
+    #     "Sec-Ch-Ua-Mobile": "?0",
+    #     "Sec-Ch-Ua-Model": '""',
+    #     "Sec-Ch-Ua-Platform": '"macOS"',
+    #     "Sec-Ch-Ua-Platform-Version": '"15.3.0"',
+    #     "Sec-Fetch-Dest": "empty",
+    #     "Sec-Fetch-Mode": "cors",
+    #     "Sec-Fetch-Site": "same-origin"
+    # }
+    # profile = request_server_detail(url, headers)
+    # if profile:
+    #     with open("server_zhipu-web-search_BigModel_profile.json", "w", encoding="utf-8") as f:
+    #         json.dump(profile, f, ensure_ascii=False, indent=2)
+    #     print("Profile saved to server_zhipu-web-search_BigModel_profile.json")
+    # else:
+    #     print("Profile extraction failed!")
+    # 
+    batch_request_servers() 

Crawler/data_cleaner.py

@@ -0,0 +1,102 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+MCP Servers Data Cleaning Script
+Keep only user-specified core fields and simplify data structure
+"""
+
+import json
+import os
+from typing import Dict, Any, List
+
+def clean_server_data(original_data: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
+    """
+    Clean server data, keep only specified fields, and filter out items without URLs
+    """
+    cleaned_data = []
+    filtered_count = 0
+    
+    for item in original_data:
+        # Filter out items without URLs
+        url = item.get('url')
+        if not url or url.strip() == '':
+            filtered_count += 1
+            continue
+            
+        metadata = item.get('metadata', {})
+        github = item.get('github', {})
+        
+        # Build cleaned data structure
+        cleaned_item = {
+            # Unique identifier
+            "id": metadata.get('id'),
+            
+            # Basic information
+            "name": item.get('name'),
+            "url": url,
+            
+            # metadata core fields
+            "title": metadata.get('title'),
+            "description": metadata.get('description'),
+            "author_name": metadata.get('author_name'),
+            "tags": metadata.get('tags'),
+            "category": metadata.get('category'),
+            "type": metadata.get('type'),
+            "tools": metadata.get('tools'),
+            "sse_url": metadata.get('sse_url'),
+            "server_command": metadata.get('server_command'),
+            "server_config": metadata.get('server_config'),
+            
+            # github complete object
+            "github": github if github else None
+        }
+        
+        cleaned_data.append(cleaned_item)
+    
+    print(f"Filtered out items without URLs: {filtered_count} records")
+    return cleaned_data
+
+def main():
+    """Main function"""
+    input_file = 'mcpso_servers.json'
+    output_file = 'mcpso_servers_cleaned.json'
+    
+    # Check input file
+    if not os.path.exists(input_file):
+        print(f"❌ Input file not found: {input_file}")
+        return
+    
+    # Read original data
+    print(f"📖 Reading original data: {input_file}")
+    with open(input_file, 'r', encoding='utf-8') as f:
+        original_data = json.load(f)
+    
+    print(f"Original data count: {len(original_data)}")
+    
+    # Clean data
+    print("🧹 Starting data cleaning...")
+    cleaned_data = clean_server_data(original_data)
+    
+    # Statistics
+    github_count = sum(1 for item in cleaned_data if item.get('github'))
+    print(f"Cleaned data count: {len(cleaned_data)}")
+    print(f"Contains GitHub information: {github_count} records")
+    
+    # Write cleaned data
+    print(f"💾 Writing cleaned data: {output_file}")
+    with open(output_file, 'w', encoding='utf-8') as f:
+        json.dump(cleaned_data, f, ensure_ascii=False, indent=2)
+    
+    # Calculate file size change
+    original_size = os.path.getsize(input_file) / 1024 / 1024  # MB
+    cleaned_size = os.path.getsize(output_file) / 1024 / 1024  # MB
+    size_reduction = (1 - cleaned_size / original_size) * 100
+    
+    print(f"\n📊 Cleaning Results:")
+    print(f"Original file size: {original_size:.2f} MB")
+    print(f"Cleaned file size: {cleaned_size:.2f} MB")
+    print(f"Size reduction: {size_reduction:.1f}%")
+    print(f"✅ Data cleaning completed!")
+
+if __name__ == "__main__":
+    main() 

Crawler/github_info_collector.py

@@ -0,0 +1,157 @@
+import os
+import re
+import sys
+import json
+import time
+import requests
+from urllib.parse import urlparse
+
+def robust_request(url, headers=None, max_retries=5, timeout=30):
+    """
+    Robust requests.get with retry logic and GitHub API rate limit handling.
+    """
+    delay = 5
+    for attempt in range(max_retries):
+        try:
+            resp = requests.get(url, headers=headers, timeout=timeout)
+            if resp.status_code == 403:
+                # Check if it's rate limit
+                if 'X-RateLimit-Remaining' in resp.headers and resp.headers['X-RateLimit-Remaining'] == '0':
+                    reset_ts = int(resp.headers.get('X-RateLimit-Reset', time.time() + 60))
+                    wait_sec = max(reset_ts - int(time.time()), 5)
+                    print(f"[Rate Limit] API rate limit, waiting {wait_sec} seconds...")
+                    time.sleep(wait_sec)
+                    continue
+                else:
+                    print(f"[WARN] 403 Forbidden: {url}")
+                    time.sleep(delay)
+                    delay = min(delay * 2, 120)
+                    continue
+            elif resp.status_code in (429, 502, 503, 504):
+                print(f"[WARN] {resp.status_code}, retrying {url}, waiting {delay} seconds...")
+                time.sleep(delay)
+                delay = min(delay * 2, 120)
+                continue
+            return resp
+        except Exception as e:
+            print(f"[ERROR] Network exception: {e}, retrying {url}, waiting {delay} seconds...")
+            time.sleep(delay)
+            delay = min(delay * 2, 120)
+    print(f"[FATAL] Multiple retries failed: {url}")
+    return None
+
+def extract_github_repo(url):
+    """Extract owner/repo from GitHub URL"""
+    if not url or 'github.com' not in url:
+        return None
+    m = re.search(r'github.com/([\w\-\.]+)/([\w\-\.]+)', url)
+    if m:
+        return f"{m.group(1)}/{m.group(2)}"
+    return None
+
+def get_github_info(full_name, token=None):
+    """Get GitHub repository information"""
+    headers = {'Accept': 'application/vnd.github+json'}
+    if token:
+        headers['Authorization'] = f'token {token}'
+    api_url = f'https://api.github.com/repos/{full_name}'
+    repo_resp = robust_request(api_url, headers)
+    if not repo_resp or repo_resp.status_code != 200:
+        print(f"[WARN] Failed to get repository info: {full_name}")
+        return None
+    repo = repo_resp.json()
+    # Contributors count
+    contrib_url = f'https://api.github.com/repos/{full_name}/contributors?per_page=1&anon=true'
+    contrib_resp = robust_request(contrib_url, headers)
+    contributors_count = 0
+    if contrib_resp and contrib_resp.status_code == 200:
+        if 'Link' in contrib_resp.headers and 'last' in contrib_resp.headers['Link']:
+            last_link = contrib_resp.headers['Link'].split(',')[-1]
+            m = re.search(r'&page=(\d+)>; rel="last"', last_link)
+            if m:
+                contributors_count = int(m.group(1))
+        else:
+            contributors_count = len(contrib_resp.json())
+    # Language statistics
+    lang_url = f'https://api.github.com/repos/{full_name}/languages'
+    lang_resp = robust_request(lang_url, headers)
+    languages = lang_resp.json() if lang_resp and lang_resp.status_code == 200 else {}
+    # File detection
+    tree_url = f'https://api.github.com/repos/{full_name}/git/trees/{repo.get('default_branch', 'main')}?recursive=1'
+    tree_resp = robust_request(tree_url, headers)
+    has_docker = has_readme = has_requirements = False
+    if tree_resp and tree_resp.status_code == 200:
+        files = [item['path'].lower() for item in tree_resp.json().get('tree', []) if item['type'] == 'blob']
+        has_docker = any('dockerfile' in f for f in files)
+        has_readme = any(f.startswith('readme') for f in files)
+        has_requirements = any('requirements.txt' in f for f in files)
+    # Last commit
+    commit_url = f'https://api.github.com/repos/{full_name}/commits?per_page=1'
+    commit_resp = robust_request(commit_url, headers)
+    last_commit = None
+    if commit_resp and commit_resp.status_code == 200 and len(commit_resp.json()) > 0:
+        last_commit = commit_resp.json()[0]['commit']['committer']['date']
+    # license
+    license_name = repo['license']['name'] if repo.get('license') else None
+    return {
+        "full_name": full_name,
+        "stargazers_count": repo.get('stargazers_count', 0),
+        "forks_count": repo.get('forks_count', 0),
+        "open_issues_count": repo.get('open_issues_count', 0),
+        "contributors_count": contributors_count,
+        "language": repo.get('language'),
+        "languages": languages,
+        "license": license_name,
+        "archived": repo.get('archived', False),
+        "has_docker": has_docker,
+        "has_readme": has_readme,
+        "has_requirements": has_requirements,
+        "last_commit": last_commit
+    }
+
+def update_json_file(json_path, token=None):
+    with open(json_path, 'r', encoding='utf-8') as f:
+        servers = json.load(f)
+    changed = False
+    for idx, item in enumerate(servers):
+        url = item.get('url')
+        if not url or 'github.com' not in url:
+            continue
+        if 'github' in item and item['github']:
+            continue  # Already collected
+        full_name = extract_github_repo(url)
+        if not full_name:
+            continue
+        print(f"[{idx+1}/{len(servers)}] Collecting {full_name} ...")
+        info = get_github_info(full_name, token)
+        if info:
+            item['github'] = info
+            changed = True
+            # Write back in real time
+            with open(json_path, 'w', encoding='utf-8') as f:
+                json.dump(servers, f, ensure_ascii=False, indent=2)
+        time.sleep(0.5)  # Prevent API rate limiting
+    if changed:
+        print(f"All collection completed, written back to {json_path}")
+    else:
+        print("No repositories need to be updated.")
+
+def test_single_url(url, token=None):
+    full_name = extract_github_repo(url)
+    if not full_name:
+        print("Not a valid GitHub URL")
+        return
+    info = get_github_info(full_name, token)
+    print(json.dumps(info, ensure_ascii=False, indent=2))
+
+if __name__ == "__main__":
+    import argparse
+    parser = argparse.ArgumentParser(description='Batch/single collection of GitHub repository information')
+    parser.add_argument('--file', type=str, default='mcpso_servers.json', help='JSON file path')
+    parser.add_argument('--url', type=str, help='Single GitHub repository URL')
+    parser.add_argument('--token', type=str, help='GitHub API Token (optional)')
+    args = parser.parse_args()
+    if args.url:
+        test_single_url(args.url, args.token)
+    else:
+        update_json_file(args.file, args.token)