CVE-2026-42073
Authentication Bypass in OpenClaude MCP Server
Publication date: 2026-06-02
Last updated on: 2026-06-02
Assigner: GitHub, Inc.
Description
Description
CVSS Scores
EPSS Scores
| Probability: | |
| Percentile: |
Meta Information
Affected Vendors & Products
| Vendor | Product | Version / Range |
|---|---|---|
| gitlawb | openclaude | to 0.5.1 (exc) |
Helpful Resources
Exploitability
| CWE ID | Description |
|---|---|
| CWE-400 | The product does not properly control the allocation and maintenance of a limited resource. |
| CWE-352 | The web application does not, or cannot, sufficiently verify whether a request was intentionally provided by the user who sent the request, which could have originated from an unauthorized actor. |
Attack-Flow Graph
AI Powered Q&A
Can you explain this vulnerability to me?
CVE-2026-42073 is a vulnerability in the OpenClaude MCP authentication flow where an attacker can bypass the OAuth state validation check due to a logic flaw in handling error parameters.
Specifically, the local OAuth callback server starts a temporary HTTP server to handle OAuth callbacks and validates a state parameter to prevent CSRF attacks. However, if an error parameter is present in the request, the state validation is skipped entirely, allowing an attacker to force the server to shut down without knowing the state value.
This flaw allows an attacker to remotely trigger a denial of service by terminating the user's active authentication session through a malicious cross-origin request.
How can this vulnerability impact me? :
This vulnerability can impact you by causing a Denial of Service (DoS) during the OAuth authentication process.
An attacker can remotely force the local OAuth callback server to shut down, terminating your active authentication session without needing any authentication or knowledge of the state parameter.
This disruption affects availability, preventing successful authentication and potentially blocking access to services relying on OpenClaude's authentication flow.
How can this vulnerability be detected on my network or system? Can you suggest some commands?
This vulnerability involves the OpenClaude MCP authentication flow where the local OAuth callback server improperly handles requests containing an error parameter, causing it to shut down unexpectedly.
To detect exploitation attempts on your system or network, monitor for unexpected shutdowns or interruptions of the OpenClaude local OAuth callback server during authentication flows.
You can also inspect HTTP requests to the local OAuth callback server for the presence of query parameters containing "error" (e.g., ?error=anything) which bypass state validation.
Suggested commands include using network monitoring or packet capture tools to filter such requests. For example, using tcpdump or tshark to capture HTTP requests to the local server port and filter for "error" parameters:
- tcpdump -i <interface> -A 'tcp port <local_oauth_port> and (((tcp[((tcp[12:1] & 0xf0) >> 2):4]) = 0x47455420))' | grep 'error='
- tshark -i <interface> -Y 'http.request.uri contains "error=" and tcp.port == <local_oauth_port>'
Replace <interface> with your network interface and <local_oauth_port> with the port used by the OpenClaude local OAuth callback server.
Additionally, reviewing OpenClaude logs for unexpected server shutdowns or error parameter handling anomalies can help detect attempts to exploit this vulnerability.
What immediate steps should I take to mitigate this vulnerability?
The primary mitigation is to upgrade OpenClaude to version 0.5.1 or later, where the vulnerability has been patched by enforcing proper OAuth state validation before processing any error parameters.
Until you can upgrade, consider restricting access to the local OAuth callback server to trusted clients only, for example by firewalling or limiting network exposure to prevent remote malicious requests.
Monitor and audit authentication flows for unexpected shutdowns or interruptions, and educate users to avoid interacting with suspicious web pages that could trigger malicious OAuth callback requests.
Review and apply any additional security controls related to sandbox permissions and trusted approvals as indicated in the v0.5.1 release notes to reduce attack surface.