/

CVE-2026-48799

Severity:

HIGH RISK

(7.7)

Improper Auth Checks (CWE-345) in Postiz Payments

CVE-2026-48799

Deferred

Vulnerability type

Category:

Improper Authentication and Authorization

CWE:

CWE-345, CWE-639

Impact:

Unauthorized Privilege and Subscription Escalation

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

Postiz contains an issue where payment callback authenticity is not properly verified against the payment provider's shared secret, and the target subscription is taken directly from untrusted request data.

Under these conditions, a low-privileged user can forge payment notifications and grant arbitrary organizations lifetime PRO subscriptions without completing a valid payment.

Affected context

This vulnerability appears in Postiz deployments that use Nowpayments or a similar payment provider for managing PRO subscriptions.

The affected code path processes Instant Payment Notification (IPN) callbacks and:

  • Does not reliably validate the callback using the provider's shared secret or signature
  • Reads the target subscription or organization identifier directly from the untrusted request body

It becomes relevant in environments where users can interact with payment-related endpoints or influence callback traffic, even indirectly.

Exploit preconditions

  • Attack surface: Public or semi-public payment notification / webhook endpoint used for Nowpayments IPN callbacks.
  • Required access: Any low-privileged Postiz account plus the ability to send crafted HTTP requests to the IPN endpoint.
  • Trigger point: Processing of a forged or manipulated payment callback that references a target organization or subscription ID.
  • Why exploitation is possible:
  • The application does not robustly verify the callback's authenticity using the payment provider's shared secret or signature.
  • The subscription or organization identifier is taken from the untrusted request body instead of a trusted mapping on the server side.
  • As a result, the system trusts attacker-controlled data to grant or modify subscription state.

Why this happens

The root cause is a combination of improper verification of cryptographic signatures (CWE-345) and insecure direct object reference (CWE-639) in the payment workflow.

The IPN handler relies on request parameters without reliably confirming that the notification originated from the legitimate payment provider using the shared secret or a validated signature.

At the same time, it accepts the target subscription or organization identifier directly from the callback payload, effectively allowing one user to reference and modify another organization's subscription state.

Abstractly, this reflects a pattern where sensitive state transitions (e.g., granting premium access) are driven by untrusted, client-supplied identifiers and insufficient message authenticity checks.

Potential impact

  • Low-privileged users may grant lifetime PRO subscriptions to arbitrary organizations without successful payment.
  • Revenue loss due to unauthorized premium feature access and subscription fraud.
  • Distortion of billing, reporting, and financial reconciliation data.
  • Erosion of trust in automated payment processing and audit trails.
  • Potential abuse of PRO-only capabilities for spam, data extraction, or platform misuse.

General mitigation guidance

Effective mitigation focuses on strengthening both authenticity checks and authorization boundaries around payment callbacks:

  • Treat all incoming payment notifications as untrusted until they are verified against the payment provider's shared secret, signature, or HMAC scheme, following the provider's recommended verification pattern.
  • Derive the target subscription or organization from trusted server-side state (e.g., an internal transaction record keyed by a secure payment reference) instead of reading identifiers directly from request bodies.
  • Ensure that subscription updates are idempotent, auditable, and driven by a well-defined state machine that only accepts transitions from validated payment events.
  • Isolate payment-handling logic into a minimal, well-reviewed component and ensure it runs with least privilege, only able to modify subscription-related fields.
  • Log all subscription upgrades and payment events with sufficient detail to support anomaly detection and post-incident analysis, without storing sensitive secrets in plaintext.

How teams detect issues like this

Detection and assurance for this vulnerability class can leverage multiple layers:

  • Static analysis and code scanning: Flag payment or webhook handlers that:
  • Do not validate request signatures or shared secrets.
  • Use user-controlled identifiers (e.g., organization_id, subscription_id) directly to update premium flags or entitlements.
  • Targeted code review: Examine payment integration modules for:
  • Clear, centralized verification of provider signatures or HMACs.
  • Server-side mapping from trusted payment transaction IDs to internal accounts, avoiding direct trust in client-supplied IDs.
  • Explicit authorization checks before modifying subscription state.
  • Runtime monitoring and logging: Monitor for patterns such as:
  • Anomalously high volume of "successful" payment callbacks from a small number of client IPs.
  • Upgrades to PRO with missing or mismatched payment records on the provider side.
  • Security testing and QA: Incorporate fuzzing and negative tests for webhook endpoints, simulating callbacks with invalid signatures, mismatched organization IDs, or altered payloads to ensure they are rejected.

Sources

SourceLink
gadvisory.orggadvisory.org/advisories/PSA-2026-Q3TCPK
github.comgithub.com/gitroomhq/postiz-app/commit/23696d2973510ae1f3f48bfa41a6bfbbf9827b05
github.comgithub.com/gitroomhq/postiz-app/releases/tag/v2.21.8
github.comgithub.com/gitroomhq/postiz-app/security/advisories/GHSA-j7rp-5mgj-qgg9

Frequently asked

questions

Yes. If the payment notification or webhook endpoint is exposed over the internet, an attacker with any valid low-privileged account and the ability to send crafted HTTP requests can attempt to exploit it remotely.

No. The vulnerability description indicates that a low-privileged account is sufficient, as the core issue lies in trusting unauthenticated or weakly authenticated payment callbacks and untrusted subscription identifiers.

Implement robust verification of all payment callbacks using the payment provider's shared secret or signature mechanisms, and drive subscription updates from server-side transaction records.

The callback should only reference an internal transaction key; the system should then look up and validate the associated organization and expected payment status before applying any subscription changes.

Adopt a standard webhook security pattern: require signature or HMAC verification, maintain a server-side mapping of pending transactions to accounts, avoid trusting client-supplied identifiers for entitlement changes, and regularly review payment-handling code for broken access control and cryptographic misuse.

Find out what's already
exploitable in your codebase.