CVE-2025-10898: Autodesk Shared Components RCE Flaw

CVE-2025-10898 Overview

CVE-2025-10898 is an out-of-bounds write vulnerability [CWE-787] affecting Autodesk Shared Components and a broad set of Autodesk 2026 products. The flaw is triggered when a maliciously crafted MODEL file is parsed by an affected application. Successful exploitation can cause a process crash, corrupt application data, or allow arbitrary code execution in the context of the current user.

Exploitation requires local file access and user interaction, since a victim must open the malicious MODEL file. The vulnerability impacts core design and engineering tools, including AutoCAD, Revit, Inventor, 3ds Max, and Civil 3D, expanding the attack surface across engineering workstations.

Critical Impact
A crafted MODEL file can trigger memory corruption in Autodesk parsers, enabling arbitrary code execution with the privileges of the user running the affected application.

Affected Products

Autodesk Shared Components
Autodesk AutoCAD 2026 and AutoCAD verticals (Architecture, Electrical, Map 3D, Mechanical, MEP, Plant 3D)
Autodesk 3ds Max 2026, Advance Steel 2026, Civil 3D 2026, InfraWorks 2026, Inventor 2026, Revit 2026, Revit LT 2026, and Vault 2026

Discovery Timeline

2025-12-16 - CVE-2025-10898 published to the National Vulnerability Database (NVD)
2025-12-19 - Last updated in NVD database

Technical Details for CVE-2025-10898

Vulnerability Analysis

The vulnerability resides in Autodesk Shared Components, a parsing library reused across the Autodesk 2026 product line. When the library processes a MODEL file, it writes data beyond the bounds of an allocated buffer. This out-of-bounds write [CWE-787] corrupts adjacent memory structures, including object metadata, function pointers, or heap control data.

Because the flaw lives in a shared parsing component, every product that consumes MODEL files inherits the same condition. A single crafted file can therefore be reused against AutoCAD, Revit, Inventor, and other listed products without modification.

Root Cause

The root cause is missing or insufficient validation of size and offset values inside the MODEL file before they are used to control write operations. The parser trusts attacker-controlled fields and writes data past the end of the destination buffer. Memory corruption of this kind frequently leads to control-flow hijacking when combined with heap grooming or use-after-free conditions.

Attack Vector

The attack vector is local and requires user interaction. An attacker delivers a crafted MODEL file through email, shared network drives, cloud collaboration platforms, or compromised project repositories. When an engineer or designer opens the file in an affected Autodesk application, the parser processes attacker-controlled data and triggers the out-of-bounds write. Code execution runs at the privilege level of the logged-in user, which on engineering workstations often includes access to design intellectual property and lateral movement opportunities.

No verified public proof-of-concept is available at the time of publication, and the issue is not listed in the CISA Known Exploited Vulnerabilities catalog. Refer to the Autodesk Security Advisory ADSK-SA-2025-0024 for vendor technical details.

Detection Methods for CVE-2025-10898

Indicators of Compromise

Unexpected crashes or WerFault.exe events involving acad.exe, revit.exe, inventor.exe, 3dsmax.exe, or other Autodesk processes immediately after opening a MODEL file.
MODEL files arriving from untrusted email senders, external file-sharing links, or unmanaged USB media.
Child processes spawned by Autodesk applications, such as cmd.exe, powershell.exe, or rundll32.exe, which are not part of normal product behavior.
Outbound network connections initiated by Autodesk processes to unfamiliar domains shortly after file open events.

Detection Strategies

Hunt for anomalous process trees where Autodesk binaries spawn script interpreters or LOLBins.
Monitor endpoint telemetry for memory access violations, heap corruption exceptions, or DEP/ASLR mitigation events originating from Autodesk processes.
Inspect file-write events that place .model files into project directories from network or removable sources prior to user open events.

Monitoring Recommendations

Enable detailed process creation and command-line auditing on engineering workstations and forward logs to a central SIEM.
Track Autodesk product versions across the fleet to confirm that patched builds are deployed.
Alert on Autodesk processes making outbound connections to non-Autodesk infrastructure or writing executable content to disk.

How to Mitigate CVE-2025-10898

Immediate Actions Required

Apply the updates listed in Autodesk Security Advisory ADSK-SA-2025-0024 to all affected 2026 products and Shared Components installations.
Use Autodesk Access to inventory installed products and push pending security updates across managed workstations.
Restrict opening of MODEL files received from external or untrusted sources until patches are confirmed in place.

Patch Information

Autodesk has released fixed builds for the affected 2026 products. Administrators should consult Autodesk Security Advisory ADSK-SA-2025-0024 for the specific patched versions per product and deploy them through Autodesk Access or the standard enterprise software distribution workflow.

Workarounds

Block inbound MODEL file attachments at the email gateway and quarantine files originating outside trusted project channels.
Enforce least-privilege user accounts on engineering workstations so that exploitation cannot escalate to administrative actions.
Validate file origin through code signing, hash allowlists, or document management systems before opening MODEL files in Autodesk applications.
Enable Windows exploit mitigations such as Data Execution Prevention (DEP), Address Space Layout Randomization (ASLR), and Control Flow Guard (CFG) for Autodesk processes.