Atomic Edge analysis of CVE-2026-1857:
The Gutenberg Blocks with AI by Kadence WP plugin contains an authenticated Server-Side Request Forgery vulnerability in versions up to and including 3.6.1. The vulnerability resides in the GetResponse REST API handler, allowing users with Contributor-level permissions to trigger arbitrary server-side requests using stored API credentials. The CVSS score of 4.3 reflects the authenticated nature and limited scope of impact.
Atomic Edge research identified the root cause in the `get_items_permission_check()` function within `/kadence-blocks/includes/advanced-form/getresponse-rest-api.php`. The function previously returned `current_user_can(‘edit_posts’)` on line 62, granting access to users with Contributor role or higher. The vulnerable `get_items()` function accepted an `endpoint` parameter without validation, allowing arbitrary path injection into the constructed API request URL on line 74. Attackers could control the `$end_point` variable appended to the base API URL.
Exploitation requires an authenticated WordPress user with at least Contributor permissions. Attackers send a POST request to the REST API endpoint `/wp-json/kadence-blocks/v1/getresponse/` with the `endpoint` parameter containing arbitrary paths. The plugin then proxies these requests to the configured GetResponse API server using stored credentials. Attackers can enumerate sensitive data like contacts, campaigns, and mailing lists. The API key leaks in request headers during these proxied calls.
The patch implements two security controls. First, the permission check in `get_items_permission_check()` now requires `current_user_can(‘manage_options’)` (Administrator only) instead of `edit_posts`. Second, the `get_items()` function validates the `endpoint` parameter against a new `ALLOWED_ENDPOINTS` constant containing only ‘campaigns’, ‘tags’, and ‘custom-fields’. Any endpoint value not in this whitelist triggers a 400 error. These changes restrict access to administrators and limit requestable paths to legitimate API resources.
Successful exploitation exposes sensitive GetResponse account data including contact lists, campaign details, and custom fields. The stored API credentials transmit with each request, potentially allowing attackers to hijack the connected GetResponse account. While the vulnerability does not enable arbitrary external SSRF, it permits unauthorized access to all data accessible via the configured GetResponse API token. This could lead to data exfiltration, privacy violations, and unauthorized use of marketing resources.







