CVE-2026-114114: Vyper source literal value substitution via multi-byte characters

CVE-2026-114114 · Severity: medium · CVSS 5.3 · Published 2026-10-21

Vyper is a smart contract programming language for Ethereum. Due to mishandled Unicode encoding, an attacker can craft source code with hidden multi-byte characters that cause numeric constants (addresses, amounts, selectors) to be silently substituted during compilation. This could result in a compiled contract containing unintended values without any compiler warning, potentially allowing theft or redirection of funds.

Technologies: Vyper. Vendors: Vyper.

Quick answers

Is CVE-2026-114114 exploited in the wild?
No. Junglewise Threat Intelligence has no report of CVE-2026-114114 being exploited in the wild.
How severe is CVE-2026-114114?
Medium severity, with a CVSS score of 5.3, published 21 October 2026.
Which products and versions are affected?
Vyper 0.1.0b1 before 0.1.0b12, 0.1.0b14, 0.4.1 through 0.5.0b1.
Is there a fix for CVE-2026-114114?
Yes: fixed in Vyper 0.1.0b12, 0.1.0b15, 0.4.0 (unaffected); patched versions after 0.5.0b1. Until then: Until upgraded, reject non-ASCII characters outside comments and docstrings, including byte-order marks, during source review or via a linter/pre-processor. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
How is CVE-2026-114114 attacked?
It is attacked over the network, in a software library.
Am I affected by CVE-2026-114114?
Check whether you run Vyper 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: Vyper 0.1.0b12, 0.1.0b15, 0.4.0 (unaffected); patched versions after 0.5.0b1

Until upgraded, reject non-ASCII characters outside comments and docstrings, including byte-order marks, during source review or via a linter/pre-processor.

Technical details

In vyper/ast/parse.py, the parser mixes character-based line offsets with Python's UTF-8 byte col_offset values. Multi-byte UTF-8 characters (such as non-ASCII identifiers that NFKC-normalize to ASCII, or an invisible byte-order mark) shift the source slice boundaries used to extract later numeric literals on the same line. An attacker can position a replacement value where the shifted slice lands, changing addresses, integers, or other constants without triggering a compiler warning. This affects versions where character offsets are converted to byte offsets, or where a line table was introduced.

Affected products

  • Vyper Vyper 0.1.0b1 before 0.1.0b12, 0.1.0b14, 0.4.1 through 0.5.0b1

Timeline

  • 2026-10-06: disclosed
  • 2026-10-21: advisory

References

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