Who is affected
Vanderbilt University's Research Electronic Data Capture (REDCap) is affected by a code injection vulnerability that allows unauthenticated remote code execution. The vulnerability requires knowledge of a valid public survey hash but does not require prior authentication. Any organisation running REDCap that exposes public surveys over the internet is at risk.
How to check whether this touches you
- Inventory: Search your infrastructure for any REDCap deployments, noting their version and whether they host public surveys (accessible without login).
- Network exposure: Confirm whether your REDCap instance is reachable from the internet or only from your internal network; test from an external IP whether you can reach the survey interface.
- Survey enumeration: List all active public survey hashes on your instance; these are the entry points an attacker would need to know.
- Version check: Query your REDCap installation to identify the running version; Vanderbilt's advisory will specify affected versions, though backported patches may exist in your deployment.
What to do
- Immediate: If you cannot patch immediately, disable public survey access at the web application firewall or load balancer level, logging all blocked requests for later review.
- Prioritise patching: Follow CISA BOD 26-04 guidance; treat this as critical if your REDCap instance is internet-facing, with a remediation window of three days from discovery.
- Logging: Enable detailed logging of HTTP requests to REDCap controller routes, file-path parameters, and stream parameters; retain logs for forensic review (see CISA Forensics Triage Requirements in the record's referenced guidance).
- Vendor communication: Consult Vanderbilt's security advisory for the precise patch version and deployment instructions; if mitigations are unavailable and you cannot patch, escalate to your security and business continuity teams to evaluate discontinuing use.
If you find you were exposed
Review access logs and REDCap audit trails from before the patch date, searching for unusual HTTP requests to controller routes with crafted file-path or stream parameters, and any unexpected code execution events. Exploitation typically precedes public disclosure, so extend your search window backwards as far as your log retention permits. If you identify suspicious activity, preserve logs, examine any data imports that occurred during the exposure window, and report findings to your incident response team for further forensic analysis.