VAITP Dataset

← Back to the dataset

CVE-2026-109527

Unbounded HTTP request body buffering leads to memory exhaustion.

  • CVSS 7.5
  • CWE-770
  • Resource Management
  • Remote

The Model Context Protocol Python SDK is the official Python SDK for Model Context Protocol servers and clients. In mcp versions before 1.29.1 and from 2.0.0a1 through 2.0.1, the Streamable HTTP session manager, SSE message endpoint, and OAuth authorization-server routes buffer complete HTTP request bodies without enforcing a maximum size. An unauthenticated remote client able to reach an affected endpoint can send a large declared or streamed request body that consumes multiple times the body size in process memory and can repeatedly exhaust memory or terminate the server. Stdio-only servers and deployments with an effective upstream request-size limit are not affected, and applications that use StreamableHTTPServerTransport directly remain responsible for enforcing their own limit. The standard SDK server applications and session managers are fixed in versions 1.29.1 and 2.1.0.

CVSS base score
7.5
Published
2026-10-14
OWASP
A04 Insecure Design
Orthogonal defect classification
Checking
Code defect classification
Missing Check
Category
Resource Management
Subcategory
Resource Exhaustion
Accessibility scope
Remote
Impact
Denial of Service (DoS)
Affected component
Model Context Protocol Python SDK
Fixed by upgrading
Yes

Solution

Upgrade the SDK to mcp ≥ 1.29.1 or ≥ 2.1.0 to apply the fix.

Vulnerable code sample

import asyncio
from mcp.http import StreamableHTTPServerTransport, Request, Response

MAX_BODY = 10 * 1024 * 1024  # 10 MB limit (for illustration)

class EchoServer:
    def __init__(self, host='127.0.0.1', port=8080):
        self.transport = StreamableHTTPServerTransport(host, port)

    async def handle(self, request: Request) -> Response:
        # VULNERABLE: reads entire request body into memory without size check
        body = await request.read()
        return Response(200, body=body)

    async def start(self):
        await self.transport.serve(self.handle)

    def run(self):
        asyncio.run(self.start())

Patched code sample

import asyncio
from mcp.http import StreamableHTTPServerTransport, Request, Response

MAX_BODY = 10 * 1024 * 1024  # 10 MB limit

class EchoServer:
    def __init__(self, host='127.0.0.1', port=8080):
        self.transport = StreamableHTTPServerTransport(host, port)

    async def handle(self, request: Request) -> Response:
        # FIX: enforce maximum request body size before buffering
        content_length = request.headers.get('Content-Length')
        if content_length is None or int(content_length) > MAX_BODY:
            return Response(413, body=b'Request Entity Too Large')
        body = await request.readexactly(int(content_length))
        return Response(200, body=body)

    async def start(self):
        await self.transport.serve(self.handle)

    def run(self):
        asyncio.run(self.start())

Published fix

Public sources link this CVE to the changes below. The code samples above were generated to illustrate the defect; these links lead to the code the project actually changed.

  • Add Streamable HTTP request body limits (#3095) (modelcontextprotocol/python-sdk)
    src/mcp/server/lowlevel/server.py, src/mcp/server/lowlevel/server.py · Server.streamable_http_app, src/mcp/server/mcpserver/server.py, src/mcp/server/mcpserver/server.py · MCPServer.run#3, src/mcp/server/mcpserver/server.py · MCPServer.run_streamable_http_async, src/mcp/server/mcpserver/server.py · MCPServer.streamable_http_app, src/mcp/server/streamable_http_manager.py, src/mcp/server/streamable_http_manager.py · StreamableHTTPASGIApp.__call__, src/mcp/server/streamable_http_manager.py · StreamableHTTPSessionManager.handle_request, src/mcp/server/streamable_http_manager.py · StreamableHTTPSessionManager.__init__
  • [v1.x] Apply the request body limit to the SSE and OAuth endpoints (#3344) (modelcontextprotocol/python-sdk)
    src/mcp/server/auth/routes.py, src/mcp/server/auth/routes.py · cors_middleware, src/mcp/server/auth/routes.py · create_auth_routes, src/mcp/server/fastmcp/server.py, src/mcp/server/fastmcp/server.py · FastMCP.sse_app, src/mcp/server/sse.py, src/mcp/server/sse.py · SseServerTransport.handle_post_message, src/mcp/server/sse.py · SseServerTransport.__init__, src/mcp/server/streamable_http_manager.py, src/mcp/server/transport_security.py
  • Apply the request body limit to the SSE and OAuth endpoints (#3336) (modelcontextprotocol/python-sdk)
    src/mcp/server/auth/routes.py, src/mcp/server/auth/routes.py · create_auth_routes, src/mcp/server/lowlevel/server.py, src/mcp/server/mcpserver/server.py, src/mcp/server/mcpserver/server.py · MCPServer.run#2, src/mcp/server/mcpserver/server.py · MCPServer.run_sse_async, src/mcp/server/mcpserver/server.py · MCPServer.sse_app, src/mcp/server/sse.py, src/mcp/server/sse.py · SseServerTransport.handle_post_message, src/mcp/server/sse.py · SseServerTransport.__init__, src/mcp/server/streamable_http_manager.py, src/mcp/server/transport_security.py
  • [v1.x] Add Streamable HTTP request body limits (#3101) (modelcontextprotocol/python-sdk)
    src/mcp/server/fastmcp/server.py, src/mcp/server/fastmcp/server.py · FastMCP.streamable_http_app, src/mcp/server/fastmcp/server.py · FastMCP.__init__, src/mcp/server/streamable_http_manager.py, src/mcp/server/streamable_http_manager.py · StreamableHTTPSessionManager.handle_request, src/mcp/server/streamable_http_manager.py · StreamableHTTPSessionManager.__init__

Cite this entry

@misc{vaitp:cve2026109527,
  title        = {{Unbounded HTTP request body buffering leads to memory exhaustion.}},
  author       = {Bogaerts, Fr\'ed\'eric and Ivaki, Naghmeh and Fonseca, Jos\'e},
  year         = {2026},
  note         = {VAITP Python Vulnerability Dataset, entry CVE-2026-109527},
  howpublished = {\url{https://netpack.pt/vaitp/vulnerability/CVE-2026-109527/}}
}

Download this entry as JSON Download the whole dataset

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 ::

Before you use the VAITP AI models

Please read and accept the following disclaimer:

By choosing "I understand and accept", you acknowledge that you understand and agree to these terms.