All articles

NewsNovember 18, 20265 min read

REDCap: unauthenticated RCE CVE-2026-90817 exploited (CISA KEV)

CISA confirms exploitation of CVE-2026-90817 in REDCap 13.3.0 and later: unauthenticated RCE using the hash of a public survey. Details and what to do now.

Researcher in a lab coat and face mask looking at coloured test tubes in a laboratory

On 18 November 2026 the US agency CISA added CVE-2026-90817 to its Known Exploited Vulnerabilities (KEV) catalogue. It is a vulnerability in REDCap (Research Electronic Data Capture), Vanderbilt University’s web platform for collecting research data. According to the National Vulnerability Database (NVD) description, the flaw allows an unauthenticated attacker to execute arbitrary code on the REDCap server. Inclusion in the catalogue means CISA has evidence of real-world exploitation.

This article is based on a single source: the CISA entry in the KEV catalogue, with NVD data. The entry does not say who is exploiting the flaw, against which organisations, since when or with which tools.

What the KEV entry says

  • Name: Vanderbilt University Research Electronic Data Capture (REDCap) Code Injection Vulnerability.
  • Added to the catalogue: 18 November 2026.
  • Deadline for US federal agencies: 21 November 2026, just three days after it was added.
  • Use in ransomware campaigns: unknown («Unknown»).
  • Required action: apply mitigations according to the vendor’s instructions, in line with CISA’s BOD 26-04 directive on prioritising security updates based on risk and its «Forensics Triage Requirements»; for cloud services follow the same directive, or stop using the product if mitigations are not available. CISA also asks organisations to assess the internet exposure of every asset.

The reference to forensic triage requirements means that updating is not enough: you also need to check whether the instance has already been compromised.

Technical details

According to the NVD, the vulnerability lies in the survey passthrough routing logic and in the Data Import processing. The attack combines two steps:

  1. by manipulating HTTP requests, the attacker reaches, from a public survey context, a controller route that should not be accessible from there;
  2. during import handling, the attacker supplies a crafted file-path or stream parameter.

If the attack succeeds, the attacker executes arbitrary code on the REDCap server. No authentication is needed, but the attacker must know the valid hash of a public survey, that is, the survey identifier. The entry does not say how an attacker obtains it; bear in mind that a public survey is designed to be opened by anyone who receives its link.

  • Weaknesses: CWE-73, external control of file name or path, and CWE-94, improper control of generation of code (code injection).
  • CVSS 3.1: 9.8 (critical), assigned by Securifera, the company whose advisories page is listed among the entry’s references. Vector AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H: network attack, low complexity, no privileges required, no user interaction, high impact on confidentiality, integrity and availability.
  • NVD entry status: published on 20 September 2026, last modified on 18 November 2026, NVD analysis still in progress («Undergoing Analysis»).

The entry has therefore been public since September: what is new is the confirmation of exploitation. In MITRE ATT&CK terms, exploiting a network-reachable instance corresponds to T1190 (Exploit Public-Facing Application). The entry does not list any public indicators of compromise.

Affected versions

The description states that the flaw exists in REDCap 13.3.0 and later versions, without naming a fixed version. Anyone running an instance from 13.3.0 onwards should consider it vulnerable until they have confirmed, through the vendor’s communications, that they have installed a version that fixes CVE-2026-90817.

What it means for European companies

REDCap collects research data through surveys, including surveys open to the public. Code execution on the server puts at risk everything the application can read: the collected data, the configuration and the database credentials. Where the platform is used for clinical studies or collects health data, these are special categories of data under the GDPR, and a compromise may require notification to the data protection authority.

The KEV catalogue is binding only on US federal agencies, but it is a widely used reference for prioritising updates. For organisations subject to NIS2 (in Italy, Legislative Decree 138/2024), including healthcare providers and research bodies in the sectors covered by the directive, an actively exploited flaw with a critical score should be handled with top priority.

What to do now

  1. Inventory: identify every REDCap instance, including those run by individual research groups or used for testing, and check their version.
  2. Update: for instances from 13.3.0 onwards, apply the fix or mitigations provided by the vendor, as CISA requires. If none are available, consider suspending the service.
  3. Exposure: assess whether the instance can be reached from the internet or from untrusted networks. Until a fix is in place, consider restricting access to the server to the networks that need it and closing public surveys that are not essential, taking into account the impact on ongoing studies.
  4. Hunt for compromise, as CISA requires with forensic triage: search web server logs for requests to public surveys with unusual parameters that reference other controllers’ routes, file paths or streams; check for new or modified files in the application folders, processes started by the web server user and unusual outbound connections.
  5. If you find suspicious signs, treat the event as an incident: updating does not remove access an attacker has already gained. Consider the database credentials and configuration secrets exposed, replace them and assess your notification obligations.

How ARXDOME helps

ARX Expose shows what can be seen of your company from outside, including exposed services and known vulnerabilities on public services: a REDCap instance reachable from the internet shows up there. On monitored servers, ARX Sentry checks file integrity in the chosen folders, which helps spot files that appear in the application folders, and ARX Trace runs collections and searches across multiple machines to check whether a compromise has taken place.

Last updated:

Any questions?

Let’s talk about your company’s needs, with no obligation.

Contact us

Related articles