CVE-2026-55620High· 7.5▾ Twilighteml_parser vulnerable to DoS via deeply nested parens in Received headers
▾ Twilight zone — High severity, or a signal on a lesser flaw
impact 41.3 · likelihood 0.1 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
Exploit-prediction probability, daily snapshots since Aug 26.
Disclosure to exploitation, from the record and what we observed since indexing it.
Disclosed via OSV
0.4%
Last analysed / modified upstream
eml_parser strips parenthesised CFWS comments from Received: headers using a regex-based fix-point loop. The loop has quadratic time complexity in the number of nested parens. A single Received: header containing 5,000 nested parens causes ~1.3 seconds of CPU saturation per parsed message; runtime quadruples per doubling of nesting depth.
This represents a CPU exhaustion DoS in any pipeline that processes attacker-supplied EML files. An attacker can create relatively small EML files that will take multiple seconds to parse.
This is particularly problematic for synchronous email-processing pipelines (gateways, sandboxes, real-time triage) where worker latency directly translates to queue backpressure and possible service-level outages.
Since version 3.0.2, eml_parser uses a linear-time algorithm to remove the comments from Received: headers.
eml-parser < 3.0.2Upgrade to a patched release:
eml-parser 3.0.2Connected by shared product, vendor, weakness, or advisory.
CVE-2026-55619Medium· 5.3eml_parser has parser DoS via deeply nested parentheses in e-mail headers
CVE-2026-55618Medium· 6.5eml_parser has a URL extraction bypass via HTML entities in URLs
CVE-2026-44844Mediumeml_parser has recursion DoS via nested message/rfc822 attachments
CVE-2026-29780Medium· 5.5eml_parser: Path Traversal in Official Example Script Leads to Arbitrary File Write
CVE-2026-39373High· 7.5JWCrypto: python-cryptography: python: JWCrypto: Memory exhaustion via crafted compressed JWE tokens (CVE-2026-39373)
CVE-2026-16100Medium· 6.5A flaw was found in the user-event metrics recording of Keycloak