CVE-2026-105208: ZITADEL IdP intent token forgery in external login

CVE-2026-105208 · Severity: high · CVSS 7.7 · Published 2026-10-04

ZITADEL is an identity management platform that authenticates users through external identity providers like Google or GitHub. The platform issues tokens to secure the external login flow, but failed to protect these tokens cryptographically. An attacker with a valid account on the same ZITADEL instance can forge tokens to hijack another user's pending login session or steal their external identity provider credentials, gaining unauthorized access to victim accounts or their connected services.

Technologies: ZITADEL. Vendors: ZITADEL.

Quick answers

Is CVE-2026-105208 exploited in the wild?
No. Junglewise Threat Intelligence has no report of CVE-2026-105208 being exploited in the wild.
How severe is CVE-2026-105208?
High severity, with a CVSS score of 7.7, published 4 October 2026.
Which products and versions are affected?
ZITADEL 4.0.0 through 4.17.2; 3.0.0 through 3.4.15.
Is there a fix for CVE-2026-105208?
Yes: fixed in ZITADEL 4.17.3 and later. Until then: Self-hosted deployments should enforce strict rate limiting and bot protection in front of the API to reduce attack success probability, though this does not fully eliminate the vulnerability. Instances not using external identity providers are unaffected as no IdP intent tokens are issued.</anigation> <parameter name="timeline">[ { "kind": "disclosed", "at": "2026-09-04", "note": "GitHub security advisory published" }, { "kind": "advisory", "at": "2026-10-04", "note": "NVD entry published" } ] Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
How is CVE-2026-105208 attacked?
It is attacked over the network, in a hosted service.
Am I affected by CVE-2026-105208?
Check whether you run ZITADEL at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

How to fix

  • Fixed in: ZITADEL 4.17.3 and later

Self-hosted deployments should enforce strict rate limiting and bot protection in front of the API to reduce attack success probability, though this does not fully eliminate the vulnerability. Instances not using external identity providers are unaffected as no IdP intent tokens are issued.</anigation> <parameter name="timeline">[ { "kind": "disclosed", "at": "2026-09-04", "note": "GitHub security advisory published" }, { "kind": "advisory", "at": "2026-10-04", "note": "NVD entry published" } ]

Technical details

ZITADEL protected IdP intent tokens using unauthenticated, malleable encryption (without an authentication tag), so tokens lacked integrity verification when redeemed. An authenticated attacker who can predict a target's in-flight intent identifier and win a timing race can tamper with their own legitimate token and replay it to impersonate another user's external login. The attack requires account access on the instance, prediction of a Sonyflake intent ID within ~10ms, race condition victory, and brute-forcing ciphertext bytes; the estimated success rate is ~1.25% within a 15-second window, mitigated further by rate limiting on hosted instances.

Affected products

  • ZITADEL ZITADEL 4.0.0 through 4.17.2; 3.0.0 through 3.4.15

References

This page as Markdown. Free to reuse under CC BY 4.0 with a link back.