CVE-2026-105748: Docling Local File Disclosure Vulnerability in Image Embedding
Summary
CVE-2026-105748: Docling: Crafted DoclingDocument JSON embeds local image files into converted output. Affects docling versions >= 2.16.0, < 2.131.0. Fixed in 2.131.0. How to fix ↓
Also tracked for
- docling-slim · PyPI
The same CVE-2026-105748 is tracked separately for each ecosystem and vendor. Its fix information is the same; the package and affected versions differ.
Check if this vulnerability affects your application
Upload a manifest or SBOM and we'll show whether you're affected, plus every other risky package in the same scan.
No signup required · Results in about 60 seconds
Background
Docling is an open-source documentation generation tool that accepts serialized DoclingDocument JSON as input. It is designed to convert documents into various formats while supporting embedded images. However, it has a vulnerability that allows crafted JSON files to reference local image files, potentially exposing sensitive information.
Vulnerability Detail
The vulnerability arises from the way Docling processes serialized DoclingDocument JSON. It does not adequately validate image references, allowing a crafted JSON file to set an image's URI to a local file path. When exporting documents with embedded images, these local files can be read and embedded as base64 in the output. This can lead to the disclosure of image files that are readable by the converting process, revealing whether a local path exists. The vulnerability is present in versions prior to 2.131.0, where the image reference validation was insufficient.
How To Fix
Latest published version: 2.142.0 as of 22 Oct 2026
pip install --upgrade "docling==2.142.0"Upgrading to it may include breaking changes; the version below is the minimum fixed release for this vulnerability.
If you need the minimum change
2.131.0
This is the first release that contains the fix.
pip install --upgrade "docling==2.131.0"Tip — Not sure which version your app runs? Vulert can match this fix to it. Scan free →
Workaround
For users unable to upgrade, consider the following workarounds: 1. Remove InputFormat.JSON_DOCLING from allowed formats when converting untrusted input. 2. Export documents using ImageRefMode.PLACEHOLDER or ImageRefMode.REFERENCED instead of EMBEDDED. Here is an example of how to use a placeholder:
from docling_core.types.doc import DoclingDocument, ImageRef
doc = DoclingDocument(name="example")
doc.add_picture(image=ImageRef(mimetype="image/png", uri="placeholder.png"))
doc.save_as_json("example.json")
Conclusion
The CVE-2026-105748 vulnerability in Docling poses a risk of local file disclosure through improperly validated image references in JSON documents. Users are advised to upgrade to the patched version or implement workarounds to mitigate the risk.
See every vulnerability recorded for docling →
FAQ
CVE-2026-105748 is a vulnerability in the Docling package that allows local file paths to be embedded in exported documents, potentially disclosing sensitive information.
Users may inadvertently expose local files when converting documents, leading to potential data leaks and unauthorized access to sensitive images.
Versions of Docling prior to 2.131.0 are affected by this vulnerability.
You can use the Vulert playground at Vulert Scanner to check if your application is affected by this vulnerability without needing to sign up.
The CVSS score of this vulnerability is 4.3, indicating a moderate severity level.
References
Related vulnerabilities
Recent vulnerabilities recorded for docling.
