CVE-2026-82286 in gpt-crawler
Summary
by MITRE • 08/28/2026
gpt-crawler through 1.5.1 fails to validate the outputFileName parameter in the POST /crawl endpoint, allowing unauthenticated attackers to write arbitrary files to any filesystem path. Attackers can supply absolute paths or parent-directory segments to overwrite existing files with content sourced from attacker-controlled URLs.
If you want to get best quality of vulnerability data, you may have to visit VulDB.
Analysis
by VulDB Data Team • 08/28/2026
The vulnerability identified in gpt-crawler versions through 1.5.1 represents a critical server-side file write flaw rooted in insufficient input validation of the outputFileName parameter within the POST /crawl endpoint. This security defect allows unauthenticated remote attackers to manipulate the destination path for files downloaded during web crawling operations. By failing to sanitize or validate the provided filename, the application permits the injection of absolute filesystem paths and directory traversal sequences such as dot-dot-slash characters. This lack of restriction enables an attacker to dictate exactly where on the server's file system the crawled content will be saved, bypassing intended sandboxing or working directories that should contain only temporary crawl data.
From a technical perspective, this flaw is classified under CWE-22 Improper Limitation of a Pathname to a Restricted Directory and CWE-436 Interpretation Conflict depending on how the underlying runtime handles path normalization versus raw string concatenation. The vulnerability stems from the application accepting user-supplied input for file naming without applying strict allow-listing or canonicalization checks before passing it to the operating system's file creation routines. When an attacker supplies a maliciously crafted filename containing absolute paths like /etc/passwd on Linux systems or C:\Windows\System32\drivers\etc\hosts on Windows, the application writes directly to those locations without verifying if they fall within acceptable boundaries. This behavior violates fundamental principles of secure coding regarding path canonicalization and directory traversal prevention as outlined in OWASP guidelines for file upload vulnerabilities.
The operational impact of this vulnerability is severe due to its potential for arbitrary file write capabilities. An attacker can overwrite critical system configuration files, web application source code, or log files with content sourced from URLs controlled by the adversary. This capability facilitates a wide range of downstream attacks including remote code execution if writable executable scripts are overwritten, privilege escalation through modification of authentication configurations, and defacement of public-facing websites hosted on the same server. Furthermore, because the vulnerability exists in an unauthenticated endpoint, it can be exploited remotely without any prior access credentials, significantly lowering the barrier to entry for malicious actors seeking to compromise the integrity of the hosting environment.
Mitigation strategies must focus on implementing robust input validation and secure file handling practices immediately upon upgrading past version 1.5.1 or applying patches if available in later releases. Developers should enforce strict allow-listing for filename characters, rejecting any input containing path separators or relative directory traversal sequences before processing. It is essential to resolve all paths to their canonical absolute forms and verify that the resulting location resides within a designated safe directory using chroot jails or containerized environments where possible. Additionally, implementing least-privilege principles by running the crawler process with minimal file system permissions can limit the damage even if an exploitation attempt succeeds. Regular security audits focusing on CWE-22 patterns in web applications are recommended to prevent similar issues from arising in other components of the software stack.