AI for Security Summit: Join Figma, Perplexity & Wiz. [Register]

CVE-2026-19583
Velociraptor vulnerability analysis and mitigation

Overview

CVE-2026-19583 is a privilege escalation vulnerability in Rapid7's Velociraptor DFIR platform that allows authenticated users with client monitoring artifact scheduling permissions to bypass access controls and execute otherwise restricted artifacts. The flaw affects all Velociraptor versions prior to 0.77.2 (Github Advisory, Feedly). It was published on September 10, 2026, with the fix merged on August 10, 2026 (GitHub PR). The vulnerability carries a CVSS v3.1 base score of 9.9 (Critical) (Github Advisory).

Technical details

The root cause is an incorrect permission assignment for a critical resource (CWE-732): Velociraptor enforces additional permission gates (e.g., the EXECVE permission) when scheduling sensitive artifacts like Linux.Sys.BashShell through normal scheduling paths, but the SetClientMonitoringState() API endpoint lacked equivalent permission validation (GitHub PR). Additionally, there was no enforcement requiring that client monitoring artifacts carry the CLIENT_EVENTS type, meaning any artifact — including those requiring elevated permissions — could be submitted as a client monitoring artifact (Github Advisory). An attacker with low-privilege authenticated access to the Velociraptor server can exploit this by calling the client monitoring scheduling API with a restricted artifact payload, bypassing the permission checks that would otherwise block the operation (Feedly).

Impact

Successful exploitation allows any authenticated user with client monitoring scheduling permissions to execute arbitrary commands on managed endpoints by scheduling restricted artifacts such as Linux.Sys.BashShell, effectively achieving remote code execution across the entire fleet of connected endpoints (Github Advisory). The scope change in the CVSS score reflects that a compromise of the Velociraptor server can cascade to all monitored endpoints, resulting in high confidentiality and integrity impact across the connected infrastructure (Feedly). This could enable lateral movement, data exfiltration, and persistent access across all systems managed by the Velociraptor deployment.

Exploitability

As of the publication date, there is no public proof-of-concept exploit and no evidence of in-the-wild exploitation (Feedly). The vulnerability requires low-level authenticated access (network-accessible, low privileges, no user interaction), making it relatively straightforward to exploit for any user with a valid Velociraptor account that has client monitoring scheduling rights. The EPSS score is 0.0, indicating low current probability of exploitation in the near term (Feedly). The CVE is not currently listed in the CISA Known Exploited Vulnerabilities catalog.

Exploitation steps

  1. Reconnaissance: Identify a Velociraptor server instance running a version prior to 0.77.2. Obtain or compromise credentials for an account with client monitoring artifact scheduling permissions (a lower-privilege role than full administrator).
  2. Identify restricted artifact: Select a high-impact artifact that normally requires elevated permissions, such as Linux.Sys.BashShell, which requires the EXECVE permission under normal scheduling.
  3. Craft malicious client monitoring request: Construct an API call to SetClientMonitoringState() that includes the restricted artifact (e.g., Linux.Sys.BashShell) in the client monitoring artifact set, without specifying the CLIENT_EVENTS type requirement.
  4. Submit the request: Send the crafted request to the Velociraptor server API. Because no permission check is enforced on the client monitoring path, the server accepts and schedules the restricted artifact.
  5. Achieve arbitrary command execution: The scheduled artifact is pushed to target endpoints and executed, resulting in arbitrary command execution on all affected managed endpoints (GitHub PR, Github Advisory).

Indicators of compromise

  • Logs: Velociraptor server audit logs showing SetClientMonitoringState() API calls from low-privilege user accounts; unexpected scheduling of artifacts such as Linux.Sys.BashShell or other artifacts requiring EXECVE permission appearing in client monitoring configurations.
  • Network: API requests to the Velociraptor server's client monitoring endpoint from accounts not typically associated with administrative artifact management.
  • File System / Endpoint: Unexpected shell processes or command execution artifacts appearing on monitored endpoints originating from the Velociraptor client agent; new files or scripts dropped by the Velociraptor agent process on managed endpoints.
  • Process: Unusual child processes spawned by the Velociraptor client agent (e.g., /bin/bash, cmd.exe) on managed endpoints that correlate with client monitoring artifact execution events.

Mitigation and workarounds

Rapid7 has released Velociraptor version 0.77.2, which enforces permission checks on the SetClientMonitoringState() API and requires that client monitoring artifacts carry the CLIENT_EVENTS type (GitHub PR, Github Advisory). Organizations should upgrade to version 0.77.2 or later immediately. As an interim measure, administrators should audit existing client monitoring artifact configurations for unauthorized or unexpected artifact entries, and restrict client monitoring scheduling permissions to only fully trusted administrator accounts until the patch is applied (Feedly).

Community reactions

The vulnerability was noted on Mastodon by the offseq security account shortly after disclosure (infosec.exchange). Coverage has been picked up by standard vulnerability aggregators including VulDB, CVEFeed, and CIRCL, indicating routine community awareness. No major vendor statements beyond the Rapid7/Velociraptor advisory and GitHub PR have been identified at this time.

Additional resources


SourceThis report was generated using AI

Related Velociraptor vulnerabilities:

CVE ID

Severity

Score

Technologies

Component name

CISA KEV exploit

Has fix

Published date

CVE-2026-19583CRITICAL9.9
  • Velociraptor logoVelociraptor
  • cpe:2.3:a:rapid7:velociraptor
NoYesSep 10, 2026
CVE-2026-19200HIGH8.9
  • Velociraptor logoVelociraptor
  • cpe:2.3:a:rapid7:velociraptor
NoYesAug 24, 2026
CVE-2026-15371HIGH8.1
  • Velociraptor logoVelociraptor
  • cpe:2.3:a:rapid7:velociraptor
NoYesAug 18, 2026
CVE-2026-19584HIGH7.7
  • Velociraptor logoVelociraptor
  • cpe:2.3:a:rapid7:velociraptor
NoYesSep 10, 2026
CVE-2026-64955MEDIUM6.1
  • Velociraptor logoVelociraptor
  • cpe:2.3:a:rapid7:velociraptor
NoYesAug 12, 2026

Free Vulnerability Assessment

Benchmark your Cloud Security Posture

Evaluate your cloud security practices across 9 security domains to benchmark your risk level and identify gaps in your defenses.

Request assessment

Get a personalized demo

Ready to see Wiz in action?

"Best User Experience I have ever seen, provides full visibility to cloud workloads."
David EstlickCISO
"Wiz provides a single pane of glass to see what is going on in our cloud environments."
Adam FletcherChief Security Officer
"We know that if Wiz identifies something as critical, it actually is."
Greg PoniatowskiHead of Threat and Vulnerability Management