{"id":"CVE-2018-20857","published":"2019-07-26T11:41:03Z","modified":"2024-08-05T12:12:29Z","title":"Zendesk Samlr before 2.6.2 allows an XML nodes comment attack such as a name_id node with user@example.com followed by…","vendor":null,"product":null,"tier":"tail","description":{"text":"Zendesk Samlr before 2.6.2 allows an XML nodes comment attack such as a name_id node with user@example.com followed by <!---->. and then the attacker's domain name.","provenance":"source","from":"CVE.org"},"cvss":null,"epss":{"score":0.01205,"percentile":0.69028,"provenance":"source"},"cwes":[],"references":[{"url":"https://github.com/zendesk/samlr/compare/v2.6.1...v2.6.2","tags":["x_refsource_MISC"]}],"priority":{"score":13.8,"factors":[{"name":"EPSS percentile","points":13.81}]},"affected":[{"ecosystem":"RubyGems","package":"samlr","vulnerableRanges":["<2.6.2"],"patched":["2.6.2"],"from":["GHSA-qpxp-5j56-gg3x"],"provenance":"source"}],"summary":{"text":"The flaw is an XML comment-injection weakness in Zendesk's Samlr, a SAML library, affecting versions before 2.6.2. Crafted XML nodes, such as a name_id value containing a legitimate-looking address followed by an embedded comment, can cause the parser to read a different identity than the one visually presented. What it does state is that an attacker can supply XML crafted this way, which could lead to identity confusion during SAML processing if the crafted content is accepted by the service.","provenance":"ai","model":"claude-haiku-5-5","generatedAt":"2026-11-20T21:53:14Z"},"euvd":{"id":"EUVD-2019-0601","exploited":false,"provenance":"source"}}