/

CVE-2026-72566

Severity:

HIGH RISK

(7.7)

SSRF (CWE-918) in automatisch HTTP Request Custom Request

CVE-2026-72566

Deferred

Vulnerability type

Category:

Server-Side Request Forgery

CWE:

CWE-918

Impact:

Unauthorized Server-Side Network Access and Data 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

A server-side request forgery vulnerability in automatisch allows an authenticated, low-privileged user with the required Flow management permission to cause the server to fetch arbitrary URLs through the HTTP Request app's Custom Request action. The full response body may be returned to the user, creating a risk of sensitive data disclosure from internal or otherwise restricted services.

Affected context

The issue occurs in automatisch deployments where:

  • The HTTP Request app is available to users who can manage Flows
  • The Custom Request action permits user-controlled request destinations
  • Server-side network access is broader than the user's intended access
  • Responses from requested endpoints are exposed through the application

Exploitation depends on the deployment's permissions, network reachability, and outbound request controls.

Exploit preconditions

  • Attack surface: The HTTP Request app's Custom Request action
  • Required access: An authenticated account with manage Flow permission
  • Trigger point: Execution of a Flow containing a user-controlled HTTP request
  • Why exploitation is possible: The server performs outbound requests without sufficiently restricting destinations and returns the resulting response body

Why this happens

The vulnerability is caused by treating a user-controlled URL as an acceptable server-side request destination without enforcing an adequate destination policy. This behavior maps to CWE-918: Server-Side Request Forgery, where an application is induced to access resources on behalf of an attacker, potentially including internal services, local interfaces, or cloud metadata endpoints.

Potential impact

  • An attacker may retrieve responses from internal or otherwise inaccessible services
  • Sensitive application, infrastructure, or cloud metadata may be disclosed
  • The server may be used to reach network locations unavailable to the attacker directly
  • Internal service endpoints may be probed or abused
  • Availability may be affected if the server is directed toward expensive or untrusted destinations

General mitigation guidance

Mitigation generally focuses on reducing trust in user-supplied destinations and restricting outbound network behavior.

Common defensive patterns include:

  • Applying strict allowlists for permitted schemes, hosts, ports, and paths
  • Resolving hostnames before connection and rejecting private, loopback, link-local, and other restricted address ranges
  • Revalidating destinations after redirects and limiting redirect behavior
  • Enforcing outbound network policies and egress controls at the runtime or infrastructure layer
  • Limiting which roles can create or execute server-side HTTP requests
  • Avoiding unnecessary exposure of complete upstream response bodies
  • Applying request timeouts, size limits, and resource controls

How teams detect issues like this

Detection approaches may include:

  • Static analysis and code review for HTTP clients that accept user-controlled URLs
  • Review of redirect handling, DNS resolution, IP-range validation, and URL parsing logic
  • Monitoring outbound requests from application workers to private, loopback, link-local, or unexpected external destinations
  • Logging request origin, destination, redirect history, and response size without recording sensitive bodies
  • Policy enforcement that requires approved destination lists and restricted egress for workflow execution
  • Automated security review tools that identify common SSRF sinks and missing destination validation

Sources

SourceLink
github.comgithub.com/automatisch/automatisch
github.comgithub.com/automatisch/automatisch/blob/main/packages/backend/src/apps/http-request/actions/custom-request/index.js

Frequently asked

questions

It may be exploitable through the application interface by an authenticated user who has the required Flow management permission, provided the server can reach the selected destination.

Not necessarily. The described condition requires a low-privileged authenticated account with manage Flow permission rather than unrestricted administrative privileges.

Restrict server-side requests to a carefully defined destination allowlist, validate resolved addresses and redirects, and enforce network-level egress controls.

Treat all user-provided URLs as untrusted, limit access to request-building features by role, avoid returning unnecessary upstream response data, and monitor outbound application traffic.

Find out what's already
exploitable in your codebase.