/

CVE-2026-86305

Severity:

HIGH RISK

(7.3)

Unrestricted Upload (CWE-434) in light0011 CMS System

CVE-2026-86305

Deferred

Vulnerability type

Category:

Unrestricted File Upload

CWE:

CWE-284, CWE-434

Impact:

Unauthorized File Upload and Potential Remote Code Execution

Severity:

HIGH RISK

(7.3)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

A remotely reachable file-upload flaw in the light0011 CMS may allow an attacker to upload files without adequate type, content, or destination restrictions. If uploaded content is stored in an executable or otherwise sensitive location, the issue could enable unauthorized code execution, website compromise, or malicious file hosting.

Affected context

The vulnerability occurs in the ThinkPHP Upload::upload function used by the CMS. Risk is increased when the upload endpoint is remotely accessible, permits insufficiently validated filenames or content, stores files beneath a web-accessible path, or allows uploaded files to be interpreted by the server. The project uses rolling releases, so affected code may be present across builds without a stable version boundary.

Exploit preconditions

  • Attack surface: A remotely accessible CMS file-upload endpoint
  • Required access: The ability to reach the upload functionality; authentication requirements depend on deployment configuration
  • Trigger point: Processing and storage of attacker-controlled uploaded content by Upload::upload
  • Why exploitation is possible: Upload restrictions or destination controls may be insufficient, allowing files that should be rejected to be accepted and made accessible to downstream web-server or application components

Why this happens

The underlying bug is an unrestricted file-upload condition mapped to CWE-434. The upload workflow appears to rely on inadequate enforcement of permitted file types, content characteristics, names, or storage destinations. When attacker-controlled files are accepted without defense-in-depth controls, the server may store content in a location where it can be executed, served as active content, or used to affect application behavior. The related CWE-284 association reflects weaknesses in authorization or access control around the upload capability.

Potential impact

  • An attacker may upload unauthorized files to the application
  • Uploaded content could be used to deface the site or host malicious resources
  • Server-side code execution could occur if executable content reaches an interpreted directory
  • Sensitive files or application data could be exposed or altered
  • The application may be used as a platform for follow-on attacks

General mitigation guidance

Mitigation generally focuses on reducing trust in uploaded content and isolating it from application execution paths.

Common defensive patterns include:

  • Allowing only narrowly defined business-required file types using server-side validation
  • Checking file content and signatures rather than trusting client-provided extensions or MIME types
  • Generating server-side filenames and rejecting traversal characters or ambiguous paths
  • Storing uploads outside web-accessible and executable directories where practical
  • Serving uploaded content with non-executable content types and restrictive response headers
  • Applying authorization and request-level controls to upload operations
  • Limiting file size, rate, quota, and total upload volume
  • Running the application with the least privileges required to write upload data

How teams detect issues like this

Organizations can look for this weakness through:

  • Static analysis rules identifying upload handlers that trust filenames, extensions, MIME types, or client-supplied paths
  • Code review checks for missing content validation, authorization, size limits, and safe storage separation
  • Dynamic testing that verifies disallowed file types and executable content are rejected
  • Web-server and application policies that prevent execution from upload directories
  • Monitoring for unusual upload volume, unexpected extensions, double extensions, path traversal indicators, and newly created executable files
  • Automated security review tools that flag unrestricted upload sinks and unsafe file-serving configurations

Sources

SourceLink
github.comgithub.com/light0011/cms/
github.comgithub.com/light0011/cms/issues/9
vuldb.comvuldb.com/cve/CVE-2026-86305
vuldb.comvuldb.com/submit/894804
vuldb.comvuldb.com/vuln/399478
vuldb.comvuldb.com/vuln/399478/cti

Frequently asked

questions

Yes, the reported attack surface is remotely reachable. Practical exploitability still depends on whether the upload endpoint is exposed and what authentication or authorization controls protect it.

Not necessarily. The required access depends on the CMS deployment and endpoint configuration; any account or session permitted to invoke the upload function may be sufficient if authorization is weak.

Validate uploads on the server, permit only required formats, generate safe names, and store content outside executable or web-accessible directories whenever possible.

Combine strict upload policies with authorization checks, content inspection, resource limits, isolated storage, non-executable serving rules, and monitoring for anomalous upload behavior.

Find out what's already
exploitable in your codebase.