UTC
High

Anthropic's MCP TypeScript SDK has a ReDoS vulnerability

CVE-2026-0621 · GHSA-8r9q-7v3j-jr4g · published 2026-01-05 · updated 2026-07-15

Severity
High
CVSS score
8.7
Chance of exploitation (EPSS)
0.44%
Fixed in
1.25.2
  • Exploitation Public proof of concept
  • Automatable Yes
  • Technical impact Partial
  • CISA SSVC
What it is

A ReDoS vulnerability in the UriTemplate class allows attackers to cause denial of service. The partToRegExp() function generates a regex pattern with nested quantifiers (([^/]+(?:,[^/]+))) for exploded template variables (e.g., {/id}, {?tags}), causing catastrophic backtracking on malicious input. Who is affected: MCP servers that register resource templates with exploded array patterns and accept requests from untrusted clients. Attack result:** An attacker sends a crafted URI via resources/read request, causing 100% CPU utilization, server hang/crash, and denial of service for all clients.

What to do

Upgrade @modelcontextprotocol/sdk to 1.25.2 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.

v1.25.2 contains b392f02ffcf37c088dbd114fedf25026ec3913d3 the fix modifies the regex pattern to prevent backtracking.

If you cannot upgrade yet

Avoid using exploded patterns ({/id}, {?tags}) in resource templates Implement request timeouts and rate limiting Validate URIs before processing to reject suspicious patterns

Affected
PackageRegistryAffectedFixed in
@modelcontextprotocol/sdknpm>= 1.3.0, < 1.25.21.25.2
Downstream in the Official MCP Registry

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

  • ai.pixeldojo/mcp@modelcontextprotocol/sdk@1.4.1
  • com.monikaskorykow/workshops@modelcontextprotocol/sdk@1.23.0
  • dev.ezstat/ezstat-mcp-server@modelcontextprotocol/sdk@1.13.2
  • io.github.anyproto/anytype-mcp@modelcontextprotocol/sdk@1.18.0
  • io.github.askman-dev/agent-never-give-up@modelcontextprotocol/sdk@1.23.0
  • io.github.AungMyoKyaw/betterprompt-mcp@modelcontextprotocol/sdk@1.18.0
  • io.github.BamaCharanChhandogi/blogcaster-mcp@modelcontextprotocol/sdk@1.20.2
  • io.github.brightdata/brightdata-mcp@modelcontextprotocol/sdk@1.21.2
  • io.github.bytedance/mcp-server-browser@modelcontextprotocol/sdk@1.15.1
  • io.github.bytedance/mcp-server-commands@modelcontextprotocol/sdk@1.15.1
  • io.github.bytedance/mcp-server-filesystem@modelcontextprotocol/sdk@1.15.1
  • io.github.bytedance/mcp-server-search@modelcontextprotocol/sdk@1.15.1
All 46 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].