cve.bz

CVE-2026-105743 docling docling: SSRF

5.8 mediumEPSS 0.2%Fix available

Medium (5.8) in docling docling. EPSS 0.2% estimated probability of exploitation activity in the next 30 days. Fixed in docling 2.132.0.

On this page
Docling simplifies document processing by parsing diverse formats and providing integrations with the generative AI ecosystem. From 2.91.0 until 2.132.0, validate_url_safety in docling/backend/utils/image_resource_loader.py validates a hostname with a single IPv4 lookup and then allows the HTTP client to resolve and parse the original URL again, permitting DNS rebinding, mixed public and internal address records, and backslash authority parser disagreement to reach internal services. HTMLBackendOptions(render_page=True) also allows HTTP and HTTPS browser requests without validating their resolved destination. Exploitation requires remote fetching to be enabled, and response content is exposed only when it is decoded as an image or passively rendered in a page screenshot. This issue is fixed in 2.132.0.

Severity

CVSS 3.1

5.8 medium
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:L/I:N/A:N
EPSS
0.2% (unchanged from 0.2% a week ago; 0.2% 30 days ago) estimated probability of exploitation activity in the next 30 days; 10th percentile. Scored 2026-11-17. A forecast for this CVE across the internet, not for your own systems.
CISA's assessment (SSVC)
Exploitation: none · Automatable: no · Technical impact: partial, assessed 2026-10-08. From CISA's Vulnrichment in the CVE List: whether it is exploited, whether exploiting it can be automated, and how much control it gives.
CWEs
CWE-367 Time-of-check Time-of-use (TOCTOU) Race Condition
CWE-918 Server-Side Request Forgery (SSRF)
OWASP Top 10
A01:2025 Broken Access Control
Published
Modified
NVD analysis
Analyzed
Record from
the NVD, synced

Fix and patched versions

Ask an AI about this CVE

Show the promptExplain CVE-2026-105743 for an engineer deciding whether to act. Use https://cve.bz/api/public/cves/CVE-2026-105743?via=ask-ai-copy and https://cve.bz/cve/CVE-2026-105743?via=ask-ai-copy as the source of truth: what is affected and which versions fix it, whether it is in CISA KEV, public exploits, and its EPSS (an internet-wide 30-day estimate of exploitation activity, not a forecast for my systems). If they give no fixed versions, use the CVE record https://cveawg.mitre.org/api/cve/CVE-2026-105743 or the vendor advisory, not third-party mirrors. Tell me how to check whether I am affected and how to verify a fix. Cite the sources and say where they are uncertain.

Opens a third-party assistant; its answer is its own, not cve.bz's. The prompt holds only this CVE's id and public links.

Affected

Versions

From the NVD's CPE matches.

ProductAffected
docling:doclingfrom 2.91.0 before 2.132.0

Packages (OSV)

AdvisoryEcosystemPackageAffectedFixed in
GHSA-pc36-qwjq-x68cPyPIdocling>=2.91.0 <2.132.02.132.0
GHSA-pc36-qwjq-x68cPyPIdocling-slim>=2.92.0 <2.132.02.132.0
PYSEC-2026-4190PyPIdocling>=2.91.0 <2.132.02.132.0
PYSEC-2026-4214PyPIdocling-slim>=2.92.0 <2.132.02.132.0

Red Hat

Red Hat rates this Low.

Red Hat's statement

A LOW severity server-side request forgery (SSRF) flaw was found in docling, a document conversion library shipped as a bundled dependency in Red Hat OpenShift AI and Lightspeed Core offerings. When remote resource fetching is enabled, the validate_url_safety check in docling's image resource loader resolves a host name once to a single IPv4 address, while the HTTP client later resolves and parses the original URL again. A remote attacker who can supply crafted URLs or untrusted documents can exploit this gap using DNS rebinding, host names with mixed public and internal address records, or URL parser confusion over backslashes in the authority, causing the converting host to send requests to internal network services such as cloud metadata endpoints and loopback services. Response content is exposed only when it decodes as an image or is rendered into a page screenshot, which limits disclosure. Red Hat rates this flaw LOW severity (CVSS 3.7), lower than the MODERATE severity assigned by NIST, because under Red Hat policy for SSRF the requests remain within the application's security authority (Scope: Unchanged) and exploitation requires high attack complexity in addition to remote fetching being enabled. The default configuration, in which remote fetching is disabled, is not affected. All affected Red Hat products are vulnerable through the bundled docling dependency; fixes are currently deferred because trackers are not filed for LOW severity flaws, and one affected stream is out of support scope. The residual risk of deferral is limited to potential partial disclosure of internal service responses in deployments that enable remote fetching and process untrusted documents.

Red Hat's mitigation

Exploitation requires remote resource fetching to be enabled, which is disabled by default in docling. Deployments that process untrusted documents should keep remote fetching disabled (enable_remote_fetch=False, avoiding CLI options such as --html-image-fetch remote or all). Where remote fetching must remain enabled, block egress from the converting host to private, link-local, and loopback address ranges at the network level, for example using firewall rules or Kubernetes network policies restricting outbound traffic from the docling service. This issue is fixed upstream in docling 2.132.0; fixes for the affected Red Hat products are deferred as described in the statement.

ProductPackageStateAdvisory
Lightspeed Corelightspeed-core/rag-tool-cpu-rhel9Fix deferred
Lightspeed Corelightspeed-core/rag-tool-cuda-12.9-rhel9Fix deferred
Red Hat OpenShift AI (RHOAI)rhai-early-access/docling-sdk-cuda-rhel9Fix deferred
Red Hat OpenShift AI (RHOAI)rhai-early-access/docling-serve-cuda-rhel9Fix deferred
Red Hat OpenShift AI (RHOAI)rhoai/odh-autorag-rhel9Fix deferred
Red Hat OpenShift AI (RHOAI)rhoai/odh-feature-server-rhel9Fix deferred
Red Hat OpenShift AI (RHOAI)rhoai/odh-ogx-core-rhel9Fix deferred

Contains Red Hat Security Data © Red Hat, Inc., licensed under CC BY 4.0. Source: https://access.redhat.com/security/data. Modified: reformatted. Red Hat's page for CVE-2026-105743.

Attack patterns (CAPEC)

  • CAPEC-664 Server Side Request Forgery Likelihood: High · Severity: High
  • CAPEC-27 Leveraging Race Conditions via Symbolic Links Likelihood: Medium · Severity: High
  • CAPEC-29 Leveraging Time-of-Check and Time-of-Use (TOCTOU) Race Conditions Likelihood: High · Severity: High

The attack patterns MITRE's CAPEC relates to this record's weaknesses, as MITRE names them. Credits.

References

Check your stackJSON