Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.
Sources: National Vulnerability Database, CISA KEV, Exploit-DB
CVE-2026-97323 Overview
CVE-2026-97323 is a path traversal vulnerability in YunaiV/zhijiantianya ruoyi-vue-pro versions up to 2026.08. The flaw resides in the getOriginalFilename function within yudao-module-mp/src/main/java/cn/iocoder/yudao/module/mp/service/material/MpMaterialServiceImpl.java, part of the File Upload component. Attackers can manipulate the filename parameter to traverse outside the intended upload directory. The attack is remotely exploitable and requires only low-level privileges. A public exploit disclosure exists, and the vendor did not respond to coordinated disclosure attempts.
Critical Impact
Authenticated remote attackers can abuse crafted filenames during material uploads to write or reference files outside the intended storage path, affecting confidentiality, integrity, and availability of the application.
Affected Products
- YunaiV/zhijiantianya ruoyi-vue-pro versions up to and including 2026.08
- The yudao-module-mp WeChat Material module
- Deployments exposing the File Upload component over the network
Discovery Timeline
- 2026-09-24 - CVE-2026-97323 published to the National Vulnerability Database (NVD)
- 2026-09-24 - Last updated in NVD database
Technical Details for CVE-2026-97323
Vulnerability Analysis
The vulnerability is a path traversal issue classified under CWE-22. It stems from unsafe handling of user-supplied filenames in the getOriginalFilename function of MpMaterialServiceImpl.java. When a client uploads WeChat material, the module uses the raw filename to construct a storage path. An attacker who submits sequences such as ../ within the filename can redirect the file write to an arbitrary location on the server filesystem. The exploit has been publicly disclosed, which lowers the barrier to opportunistic abuse.
Root Cause
The root cause is missing sanitization and canonicalization of the filename returned by getOriginalFilename. The application trusts the client-supplied name without stripping directory separators or validating that the resolved path remains within the designated upload directory.
Attack Vector
An authenticated remote attacker with upload privileges submits a multipart upload request containing a crafted filename field that embeds relative path segments. The server resolves the path against the material storage directory, writing the file outside the intended location. Depending on filesystem permissions, this can lead to overwriting application resources, planting web-accessible content, or exposing sensitive files. No verified public exploit code is included here. See the VulDB entry for CVE-2026-97323 for additional technical context.
Detection Methods for CVE-2026-97323
Indicators of Compromise
- Multipart upload requests to WeChat Material endpoints containing ../, ..\, URL-encoded %2e%2e%2f, or absolute path segments in the filename parameter.
- Files appearing outside the configured upload directory with timestamps matching upload activity.
- Unexpected modifications to static web resources, configuration files, or templates served by the application.
Detection Strategies
- Inspect HTTP request logs for upload endpoints exposed by the yudao-module-mp component and alert on filenames containing traversal sequences.
- Enable web application firewall (WAF) rules that normalize and reject path traversal payloads in multipart form fields.
- Correlate application-level upload events with filesystem write events outside the designated storage directory.
Monitoring Recommendations
- Monitor filesystem integrity on directories adjacent to the configured material upload path.
- Audit authenticated user sessions that perform material uploads and flag anomalous volumes or filename patterns.
- Capture and retain full multipart request bodies for upload endpoints to support forensic review.
How to Mitigate CVE-2026-97323
Immediate Actions Required
- Restrict network access to the ruoyi-vue-pro admin and material upload endpoints to trusted users only.
- Revoke or tightly scope accounts holding material upload privileges until a patched build is deployed.
- Deploy WAF signatures that normalize and block path traversal sequences in multipart upload parameters.
Patch Information
The vendor was contacted but did not respond to the disclosure, and no official fixed version is referenced in the NVD entry at the time of publication. Monitor the upstream ruoyi-vue-pro repository for patched releases and apply updates once available. Until then, apply the input validation workaround below.
Workarounds
- Patch the getOriginalFilename logic locally to strip directory separators and reject names containing .. or absolute paths before path construction.
- Resolve the final upload path with a canonicalization check (for example, Path.toRealPath()) and verify it starts with the configured storage base directory.
- Store uploaded files under server-generated identifiers rather than client-supplied names, preserving the original name only as metadata.
# Example input validation hardening (pseudocode)
original = multipartFile.getOriginalFilename()
if original contains any of ["..", "/", "\\", ":"]:
reject upload
safeName = UUID.randomUUID() + extensionOf(original)
target = baseDir.resolve(safeName).toRealPath()
if not target.startsWith(baseDir.toRealPath()):
reject upload
Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.
Sources: National Vulnerability Database, CISA KEV, Exploit-DB

