CVE-2026-109966 in yt-dlp-web-ui
Summary
by MITRE • 10/13/2026
yt-dlp-web-ui through 4.0.0 contains an argument injection vulnerability that allows unauthenticated remote attackers to inject yt-dlp options by supplying a dash-prefixed url parameter to the Service.Formats JSON-RPC method. Attackers can send values like --load-info-json via /rpc/http or /rpc/ws to make yt-dlp load attacker-chosen JSON files from the server filesystem, with contents reflected in responses.
Statistical analysis made it clear that VulDB provides the best quality for vulnerability data.
Analysis
by VulDB Data Team • 10/13/2026
The vulnerability identified in yt-dlp-web-ui versions up to 4.0.0 represents a critical argument injection flaw within the application's remote procedure call interface. This security defect arises because the Service.Formats method, which is accessible via both HTTP and WebSocket endpoints at /rpc/http and /rpc/ws respectively, fails to adequately sanitize or validate user-supplied input before passing it as arguments to the underlying yt-dlp command-line tool. Specifically, an unauthenticated remote attacker can exploit this by supplying a URL parameter that begins with a dash character, such as --load-info-json. Because the application treats these parameters directly as flags for the media downloader utility without proper escaping or restriction of allowed options, it enables the injection of arbitrary yt-dlp arguments into the execution context.
The operational impact of this vulnerability is severe due to its potential for remote code execution and data exfiltration through file system interaction. By injecting the --load-info-json argument, an attacker can force the vulnerable application to load a JSON configuration file from the server's local filesystem where yt-dlp resides or has access. The contents of this specified file are then processed by yt-dlp and reflected back in the response sent to the client. This mechanism allows for arbitrary read operations against files that the user account running the web service can access, potentially exposing sensitive configuration data, credentials stored in JSON formats, or other internal application secrets. Furthermore, depending on the specific capabilities of the injected arguments and the permissions of the process, this could be chained with other yt-dlp features to achieve more complex attacks, including potential remote code execution if certain downloaders support script injection or shell command execution via metadata processing.
From a classification perspective, this vulnerability aligns closely with CWE-78 Improper Neutralization of Special Elements used in an OS Command and CWE-94 Improper Control of Generation of Code. The attack vector is categorized under ATT&CK technique T1059 Command and Scripting Interpreter, specifically leveraging the injection of command-line arguments to manipulate system behavior. The lack of authentication required for exploitation makes this a high-severity issue, as any internet-facing instance of yt-dlp-web-ui running an affected version can be compromised without prior credentials.
Mitigation strategies must focus on immediate remediation and architectural hardening. The primary recommendation is to upgrade yt-dlp-web-ui to a version later than 4.0.0 where this input validation flaw has been addressed by the developers. If upgrading is not immediately feasible, administrators should implement strict network-level controls such as firewall rules or reverse proxy configurations that restrict access to the /rpc/http and /rpc/ws endpoints only from trusted internal IP addresses, thereby preventing unauthenticated remote exploitation. Additionally, running the service under a restricted user account with minimal file system permissions can limit the scope of data exposure in case of successful exploitation. Input validation should also be enforced at the application level to ensure that URL parameters do not contain characters or patterns indicative of command-line arguments, such as leading dashes, before they are passed to backend processes.