Description

Plate is a rich-text editor with AI and shadcn/ui. Prior to 53.3.2, @platejs/docx-io fetches remote image URLs while converting attacker-controlled HTML through htmlToDocxBlob in a server-side or privileged environment. The converter can make requests to internal network resources and include the fetched image bytes in the generated DOCX, allowing server-side request forgery with response disclosure. Applications can also incur resource consumption from attacker-selected remote responses. This issue is fixed in version 53.3.2.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "8.2",
  "pubDate": "2026-08-20T17:19:23.433Z",
  "pubdate": "2026-08-20T17:19:23.433Z",
  "executiveSummary": "A Server-Side Request Forgery (SSRF) vulnerability exists in the @platejs/docx-io component of Plate prior to version 53.3.2. The vulnerability arises from insecure handling of remote image URLs when converting attacker-controlled HTML via the htmlToDocxBlob function in server-side or privileged environments. This flaw allows an unauthenticated remote attacker to induce the server into making arbitrary outbound HTTP requests to internal network resources. Consequently, attackers can achieve response disclosure by forcing the inclusion of fetched image bytes directly into the generated DOCX file output. Furthermore, the vulnerability exposes applications to excessive resource consumption risks driven by attacker-selected remote responses. Successful exploitation requires the application to process untrusted HTML inputs containing maliciously crafted remote image references within a vulnerable server-side context. The high-risk implications include potential exposure of internal network infrastructure, unauthorized data exfiltration via document generation, and denial of service through resource exhaustion. Remediation requires upgrading the affected @platejs/docx-io package to version 53.3.2 or later.",
  "technicalDetails": "The vulnerability is localized within the htmlToDocxBlob function provided by the @platejs/docx-io package, which is part of the Plate rich-text editor ecosystem. The root cause of the vulnerability is the absence of input validation, URL sanitization, and domain restriction when processing image sources embedded within attacker-controlled HTML strings during the document conversion process. When executed in a server-side or privileged environment, the vulnerable component automatically attempts to fetch resources specified in image tags using remote URLs without verifying the destination against an allowed list or blocking access to private and internal IP address spaces.\nThe attack flow proceeds as follows: First, an attacker crafts a malicious HTML payload containing image elements with uniform resource locators (URLs) pointing either to internal network resources (such as loopback addresses, local metadata services, or private subnet infrastructure) or to external attacker-controlled endpoints designed to consume excessive server resources. Second, the victim application passes this attacker-controlled HTML payload to the htmlToDocxBlob function for server-side processing and conversion into a DOCX binary blob. Third, during the conversion lifecycle, the vulnerable component initiates outbound HTTP requests to fetch the specified image bytes from the provided URLs. Fourth, the server processes the responses, embedding the retrieved image bytes directly into the resulting DOCX document structure. Finally, the generated document is returned to the user or stored, disclosing the contents of the fetched internal resource to the attacker through the embedded image bytes.\nThe vulnerable component is explicitly @platejs/docx-io in versions prior to 53.3.2. Exploitation does not require authentication or specific user privileges within the application itself, provided the attacker can supply untrusted HTML content to the server-side conversion pipeline. Network exposure depends on the deployment architecture, but internal network access is inherently granted to the host executing the vulnerable server-side code. Post-exploitation impact encompasses unauthorized internal network reconnaissance, access to internal web services behind firewalls, data exfiltration of internal resources via document response disclosure, and application-level resource exhaustion."
}