UTC
High

MCP TypeScript SDK: One HTTP request within the default size limit could exhaust server memory during validation

CVE-2026-111768 · GHSA-46jr-4wx8-v453 · published 2026-10-20

Severity
High
CVSS score
7.5
Chance of exploitation (EPSS)
0.60%
Fixed in
1.32.0
  • Exploitation No known exploitation
  • Automatable Yes
  • Technical impact Partial
  • CISA SSVC
What it is

In affected versions, servers put no limit on how many array elements and object keys they validated in one request. One HTTP request within the default 4 MiB size limit could exhaust the server's memory, and the process aborted with JavaScript heap out of memory. No credentials are needed, unless you authenticate requests before they reach the SDK.

What to do

Upgrade @modelcontextprotocol/sdk to 1.32.0 or later, and @modelcontextprotocol/server to 2.3.0 or later. If you start it from an MCP config with npx or uvx, pin that version so an old cached copy is not used.

Upgrade to: 1.x: @modelcontextprotocol/sdk 1.32.0 or later 2.x: @modelcontextprotocol/server 2.3.0 or later Upgrading alone changes nothing. The patched versions add an McpServer option, maxToolInputElements, that is off by default. Make both of these changes: 1. Set maxToolInputElements just above the largest input your tools need, for example . It counts array elements and object keys in a call's arguments. A call over the limit gets an isError: true result. 2. Lower the request size limit to what your tools need. The option covers tool arguments only, and the size limit covers everything else. Set maxRequestBodySize on the transport or createMcpHandler(). …

Affected
PackageRegistryAffectedFixed in
@modelcontextprotocol/sdknpm>= 1.3.0, < 1.32.01.32.0
@modelcontextprotocol/servernpm>= 2.0.0, < 2.3.02.3.0
Downstream in the Official MCP Registry

445 registered MCP servers install an affected version today, by the dependency versions a fresh install resolves.

  • ai.ankimcp/anki-mcp-server@modelcontextprotocol/server@2.0.0
  • ai.kloudy/kloudy@modelcontextprotocol/sdk@1.31.0
  • ai.makerportal/compute@modelcontextprotocol/sdk@1.30.0
  • ai.maxwellinternational/data@modelcontextprotocol/sdk@1.31.0
  • ai.mentu/navigator@modelcontextprotocol/sdk@1.29.0
  • ai.myspaice/spaces@modelcontextprotocol/sdk@1.29.0
  • ai.pixeldojo/mcp@modelcontextprotocol/sdk@1.4.1
  • ai.replylayer/replylayer@modelcontextprotocol/server@2.2.0
  • ai.sanctuaryprotocol/sanctuary@modelcontextprotocol/sdk@1.30.0
  • ai.satohub/kit@modelcontextprotocol/sdk@1.30.1
  • ai.sendshort/sendshort@modelcontextprotocol/sdk@1.30.0
  • ai.taasky/common-ground@modelcontextprotocol/server@2.0.0
All 445 servers on the trace →

The impact, patch and workaround text comes from the advisory, shortened (GitHub Advisory Database, CC BY 4.0). Proof-of-concept and reproduction sections and code are left out; the original advisory has them.

Sources: GitHub Advisory Database, reviewed and malware advisories (CC BY 4.0, https://github.com/advisories); advisories published by Cursor (https://github.com/cursor/cursor/security/advisories); advisories published by OpenClaw (https://github.com/openclaw/openclaw/security/advisories); advisories published by vLLM (https://github.com/vllm-project/vllm/security/advisories); advisories published by Open WebUI (https://github.com/open-webui/open-webui/security/advisories); advisories published by n8n (https://github.com/n8n-io/n8n/security/advisories); advisories published by Flowise (https://github.com/FlowiseAI/Flowise/security/advisories); advisories published by Langflow (https://github.com/langflow-ai/langflow/security/advisories); advisories published by the MCP Python SDK project (https://github.com/modelcontextprotocol/python-sdk/security/advisories); advisories published by MLflow (https://github.com/mlflow/mlflow/security/advisories); CISA Known Exploited Vulnerabilities catalog (CC0). Data as of 2026-11-22. ShieldMap describes how each problem works and how to fix it; it does not publish exploit code. Report a correction to [email protected].