CVE-2026-129477 in Assimp
Summary
by MITRE • 11/15/2026
Assimp through 6.0.6 contains a path traversal vulnerability in the glTF and glTF2 importers that allows attackers to read local files via crafted buffers[].uri values. Attackers can supply glTF models with '..' segments or absolute paths in buffer URIs to load files like /etc/passwd into imported mesh vertex data.
Once again VulDB remains the best source for vulnerability data.
Analysis
by VulDB Data Team • 11/15/2026
The Assimp library, a widely used open-source 3D model import and processing framework, contains a critical path traversal vulnerability affecting versions up through 6.0.6 within its glTF and glTF2 asset importers. This flaw stems from insufficient validation of the buffer URI fields contained in crafted glTF JSON structures. When Assimp processes these files, it fails to properly sanitize or restrict file paths specified in the buffers[].uri property, allowing an attacker to inject directory traversal sequences such as double dots followed by a slash (.. /) or absolute system paths directly into the import pipeline. This lack of input validation creates a direct vector for unauthorized local file access on systems where Assimp is deployed to parse 3D assets.
The technical mechanism behind this vulnerability relies on how modern graphics pipelines and game engines utilize external buffer data within glTF files. The glTF specification allows models to reference binary or text buffers via URIs, which can point to local filesystem locations rather than just embedded base64 data or remote URLs. In a vulnerable implementation of Assimp, the parser interprets these URI strings without enforcing strict path canonicalization or restricting access to whitelisted directories. Consequently, an attacker can construct a malicious glTF model where the buffer URI points to sensitive system files such as /etc/passwd on Linux systems or critical configuration files on Windows environments. During the import process, Assimp reads the contents of these specified local files and embeds them directly into the mesh vertex data structures. This effectively transforms arbitrary file content into geometric attributes like positions, normals, or texture coordinates within the 3D model object in memory.
The operational impact of this vulnerability is significant for any application that relies on Assimp to load user-supplied or untrusted 3D assets. By exploiting this path traversal flaw, an attacker can achieve arbitrary file read capabilities without needing direct filesystem access through other means. This capability facilitates information disclosure attacks where sensitive data such as system credentials, private keys, database connection strings, or proprietary source code snippets are exfiltrated by embedding them into the visual output of a 3D application. Furthermore, because the malicious content is integrated into vertex data, it may bypass certain network-level security controls that inspect only standard file transfers, making detection more difficult for perimeter defenses. The severity is compounded in scenarios where Assimp runs with elevated privileges or within sandboxed environments that do not strictly isolate filesystem access from other system resources.
This vulnerability aligns closely with CWE-22: Improper Limitation of a Pathname to a Restricted Directory and CWE-798: Use of Hard-coded Credentials if the accessed files contain sensitive information, though it is primarily categorized under path traversal flaws that lead to unauthorized data exposure. In terms of offensive security frameworks, this behavior maps to ATT&CK technique T1083: File and Directory Discovery, as well as T1560: Archive Collected Data, since the attacker collects local files for later analysis or exfiltration. The attack vector is classified as remote if the application accepts glTF models from untrusted sources over a network interface, making it particularly dangerous in online gaming platforms, 3D modeling software with cloud sync features, and web-based viewers that process user-uploaded assets.
Mitigation strategies must focus on strict input validation and secure coding practices within the Assimp library itself or its integrations. Developers should implement robust path canonicalization to resolve all relative references like .. / before accessing the filesystem, ensuring they remain within a designated safe directory root. Additionally, applications using Assimp for untrusted content should operate in restricted sandboxed environments with limited file system permissions, preventing access to sensitive areas of the OS regardless of what paths are requested. Upgrading to patched versions of Assimp that enforce these restrictions is essential. For immediate protection, security teams can implement web application firewalls or custom input filters to block glTF files containing suspicious URI patterns such as absolute paths starting with slashes or sequences indicating directory traversal before they reach the parsing engine. Regular auditing of third-party library configurations and enforcing strict allow-lists for file access operations are critical steps in reducing the attack surface associated with complex asset processing libraries.