/

CVE-2026-53549

Severity:

HIGH RISK

(7.7)

SSRF (CWE-918) in Termix Proxy Testing Endpoint API

CVE-2026-53549

Deferred

Vulnerability type

Category:

Server-Side Request Forgery

CWE:

CWE-918

Impact:

Internal Network Access and Sensitive Information Disclosure

Severity:

HIGH RISK

(7.7)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

Termix contains a server-side request forgery vulnerability in its proxy connectivity testing endpoint. An authenticated user may cause the server to connect to attacker-selected destinations, potentially probing internal services or retrieving sensitive metadata when those destinations are reachable from the Termix host.

Affected context

The issue occurs in Termix deployments that expose the proxy testing functionality and accept destination details through the affected request fields without adequately restricting target addresses. Risk depends on the network access available to the Termix server, including access to localhost services, private networks, link-local endpoints, and cloud metadata services.

Exploit preconditions

  • Attack surface: The authenticated proxy connectivity testing endpoint
  • Required access: A valid Termix account with permission to invoke the endpoint
  • Attacker control: The ability to provide proxy and test-target host and port values
  • Trigger point: Raw TCP or SOCKS connectivity testing performed by the Termix server
  • Why exploitation is possible: Destination addresses are not sufficiently validated or restricted before outbound connections are attempted, while connection results can reveal reachability and timing information

Why this happens

The vulnerability results from treating user-controlled network destinations as trusted connection targets. The proxy testing logic performs server-side TCP and SOCKS connections without enforcing an appropriate destination policy, allowing requests to reach internal or otherwise restricted address ranges. This behavior maps to CWE-918: Server-Side Request Forgery, where an application makes network requests to attacker-influenced destinations from a trusted server context.

Potential impact

  • An attacker may enumerate localhost and private network services
  • The vulnerable server could be used to reach link-local or cloud metadata endpoints
  • Structured errors may disclose host reachability, service behavior, and timing information
  • Successful metadata access could expose cloud credentials or temporary tokens
  • Internal administrative services could be accessed or probed from the Termix network position
  • The vulnerability could support further compromise of connected infrastructure

General mitigation guidance

Mitigation generally focuses on reducing trust in user-supplied destinations and applying an explicit outbound network policy.

Common defensive patterns include:

  • Permit only explicitly approved proxy and test destinations where feasible
  • Reject loopback, private, link-local, multicast, reserved, and other non-routable address ranges unless business requirements justify them
  • Resolve hostnames and validate the resulting addresses before connecting, while accounting for DNS rebinding and resolution changes
  • Restrict outbound traffic from the Termix runtime using network segmentation and egress controls
  • Avoid exposing detailed connection errors, timing differences, or response content to users
  • Run the service with minimal network privileges and isolate access to cloud metadata services
  • Apply authorization checks so connectivity testing is available only to trusted roles

How teams detect issues like this

Organizations can detect SSRF risks through:

  • Static analysis rules identifying user-controlled host and port values passed to TCP, HTTP, SOCKS, or socket APIs
  • Code review checks for missing validation of loopback, private, link-local, and cloud metadata address ranges
  • Tests that verify redirects, DNS resolution, IPv4 and IPv6 representations, and hostname-to-address changes are constrained
  • Egress firewall and proxy logs showing application connections to internal ranges or metadata endpoints
  • Runtime monitoring for unusual destination ports, repeated connectivity probes, and access attempts from the Termix service identity
  • Policy enforcement requiring allowlisted outbound destinations for server-side connection features

Sources

SourceLink
github.comgithub.com/Termix-SSH/Termix/commit/52f4e51ae03b5b8d2608e1383e2ccf79d290132b
github.comgithub.com/Termix-SSH/Termix/pull/874
github.comgithub.com/Termix-SSH/Termix/releases/tag/release-2.3.2-tag
github.comgithub.com/Termix-SSH/Termix/security/advisories/GHSA-x9pr-795g-rm5f

Frequently asked

questions

It can be exercised through the web application by an authenticated user, provided that the user can access the proxy testing functionality and the Termix server can reach the selected destination.

The attacker generally needs valid authentication and sufficient authorization to submit proxy connectivity tests. The impact also depends on the server's network placement and egress permissions.

Treat all destination fields as untrusted, enforce an explicit allowlist or strict address policy, and block access to internal and metadata networks unless specifically required.

Combine destination validation with DNS-aware checks, network-level egress restrictions, least-privilege service accounts, and monitoring for unexpected outbound connections or detailed connectivity disclosures.

Find out what's already
exploitable in your codebase.