VAITP Dataset

← Back to the dataset

CVE-2024-53526

Command Execution in composio>=0.5.40 via handle_tool_calls.

  • CVSS 6.4
  • CWE-77
  • Input Validation and Sanitization
  • Remote

composio >=0.5.40 is vulnerable to Command Execution in composio_openai, composio_claude, and composio_julep via the handle_tool_calls function.

CVSS base score
6.4
Published
2025-01-08
OWASP
A03 Injection
Orthogonal defect classification
Algorithm
Code defect classification
Incorrect Functionality
Category
Input Validation and Sanitization
Subcategory
Command Injection
Accessibility scope
Remote
Impact
Arbitrary Code Execution
Affected component
composio
Fixed by upgrading
Yes

Solution

Upgrade to composio >= 0.5.45

Vulnerable code sample

import subprocess
import json

def handle_tool_calls(tool_calls):
    """Vulnerable function that demonstrates the security issue."""
    for tool_call in tool_calls:
        function_name = tool_call.get("function", {}).get("name")
        arguments = tool_call.get("function", {}).get("arguments")
        if function_name == "run_shell_command":
            if arguments:
                try:
                    arguments_dict = json.loads(arguments)
                    command = arguments_dict.get("command")
                    if command:
                      process = subprocess.Popen(command, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
                      stdout, stderr = process.communicate()
                      return_code = process.returncode
                      return {
                         "stdout": stdout.decode("utf-8"),
                         "stderr": stderr.decode("utf-8"),
                         "return_code": return_code
                      }
                except json.JSONDecodeError:
                     return "Invalid JSON arguments"
    return "No tool calls"

if __name__ == '__main__':
  tool_call_example = [
    {
      "id": "1",
      "function": {
        "name": "run_shell_command",
        "arguments": '{"command": "ls -la && whoami"}'
       }
    }
  ]
  
  result = handle_tool_calls(tool_call_example)
  print(result)

Patched code sample

import json
import subprocess
import shlex

def handle_tool_calls(tool_call):
    """Secure function that fixes the vulnerability."""
    
    if tool_call["function"]["name"] == "execute_command":
        arguments = tool_call["function"]["arguments"]

        try:
            parsed_arguments = json.loads(arguments)
            command = parsed_arguments.get("command", "")
            if not command:
                return "Error: Command cannot be empty"

            escaped_command = shlex.split(command)

            result = subprocess.run(escaped_command, capture_output=True, text=True, check=False)
            return result.stdout if result.returncode == 0 else result.stderr
        except json.JSONDecodeError:
            return "Error: Invalid JSON format in arguments"
        except Exception as e:
            return f"Error: {str(e)}"
    else:
       return "Unknown function name"

if __name__ == '__main__':
    
    tool_call_example = [
        {
        "id": "1",
        "function": {
            "name": "run_shell_command",
            "arguments": '{"command": "ls -la && whoami"}'
        }
        }
    ]
  
    result = handle_tool_calls(tool_call_example)
    print(result)

Payload

{"tool_calls": [{"function": {"name": "any_name", "arguments": '{"command": "whoami"}'}}]}

Cite this entry

@misc{vaitp:cve202453526,
  title        = {{Command Execution in composio>=0.5.40 via handle_tool_calls.
}},
  author       = {Bogaerts, Fr\'ed\'eric and Ivaki, Naghmeh and Fonseca, Jos\'e},
  year         = {2025},
  note         = {VAITP Python Vulnerability Dataset, entry CVE-2024-53526},
  howpublished = {\url{https://netpack.pt/vaitp/vulnerability/CVE-2024-53526/}}
}
Introducing the "VAITP dataset": a specialized repository of Python vulnerabilities and patches, meticulously compiled for the use of the security research community. As Python's prominence grows, understanding and addressing potential security vulnerabilities become crucial. Crafted by and for the cybersecurity community, this dataset offers a valuable resource for researchers, analysts, and developers to analyze and mitigate the security risks associated with Python. Through the comprehensive exploration of vulnerabilities and corresponding patches, the VAITP dataset fosters a safer and more resilient Python ecosystem, encouraging collaborative advancements in programming security.

The supreme art of war is to subdue the enemy without fighting.

Sun Tzu – “The Art of War”

:: Shaping the future through research and ingenuity ::