Executive brief
Mint is a low-level HTTP client for the Elixir programming language. A vulnerability in how it handles request methods allows an attacker to inject malicious commands into a web request if the application uses untrusted input to define the HTTP method. This could allow an attacker to bypass security controls, forge headers, or perform unauthorized actions on internal servers.
Technical details
The vulnerability exists in Mint's HTTP/1 request encoder, specifically within the `encode_request_line/2` function in `lib/mint/http1/request.ex`. The library writes the user-supplied `method` and `target` strings directly to the socket without validating them for Carriage Return or Line Feed (CRLF) characters. If an application passes attacker-controlled input as the HTTP method, the attacker can terminate the request line early and inject arbitrary headers or even a second, 'smuggled' HTTP request. While version 1.7.0 introduced validation for the `target` field, the `method` field remained unvalidated until version 1.9.0. This can lead to request smuggling and cache poisoning when Mint is used in proxy or gateway applications.
Affected products
- elixir-mint mint < 1.9.0
Timeline
- 2026-06-02: disclosed: NVD publication date
- 2026-07-09: advisory: GitHub Advisory published