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-2025-6623 Overview
CVE-2025-6623 is a stored cross-site scripting (XSS) vulnerability in the WSO2 carbon console. The console accepts user-supplied XML input without sufficiently encoding potentially executable content. An authenticated attacker on an adjacent network can inject malicious scripts that execute in a victim's browser session. Successful exploitation allows arbitrary JavaScript execution, which attackers can abuse to redirect users to malicious sites or modify the rendered user interface. Session hijacking is not feasible because sensitive session cookies carry the httpOnly flag. The flaw is classified under CWE-79 and requires user interaction to trigger.
Critical Impact
Authenticated adjacent attackers can execute arbitrary JavaScript in a victim's carbon console session, enabling UI tampering and malicious redirection.
Affected Products
- WSO2 carbon console (as identified in WSO2 Security Advisory WSO2-2025-4316)
- WSO2 products built on the carbon-analytics-common component
- Deployments exposing the carbon management console to adjacent-network users
Discovery Timeline
- 2026-10-19 - CVE-2025-6623 published to the National Vulnerability Database (NVD)
- 2026-10-20 - Last updated in NVD database
Technical Details for CVE-2025-6623
Vulnerability Analysis
The carbon console processes XML submitted by authenticated users and renders content derived from that input back into console pages. The server does not sufficiently encode potentially executable content in the XML, which allows script payloads to survive into the rendered DOM. When another authenticated user loads an affected view, the browser parses the injected markup and executes attacker-controlled JavaScript.
Exploitation requires adjacent-network access, high privileges, and user interaction. These prerequisites reduce the attack surface compared to unauthenticated or network-reachable XSS flaws. The scope is marked as changed because script execution crosses from the server-rendered response into the victim browser's security context, affecting resources the victim can reach within the console.
Impact is limited to confidentiality and integrity of in-browser content. The httpOnly attribute on session cookies prevents direct cookie theft through document.cookie, which blocks straightforward session hijacking paths.
Root Cause
The root cause is improper neutralization of input during web page generation [CWE-79]. The carbon console accepts XML, extracts fields, and reflects or stores them for later rendering without applying context-aware output encoding. Markup characters such as <, >, and quoted attributes retain their executable meaning when inserted into HTML sinks, allowing <script> and event-handler payloads to execute.
Attack Vector
An attacker with high-privilege console access submits crafted XML through a management function that stores or echoes the data. A second user, typically an administrator reviewing the same resource, loads the affected page and triggers script execution. Payloads can rewrite console UI elements, inject fake login prompts, or navigate the victim to attacker-controlled URLs. The adjacent-network requirement restricts the attack to actors reachable on the same logical network segment as the console.
No verified public exploit or proof-of-concept code is available at the time of writing. The upstream code change is tracked in the WSO2 carbon-analytics-common pull request #870.
Detection Methods for CVE-2025-6623
Indicators of Compromise
- Stored XML entries in carbon console configuration or analytics artifacts containing <script> tags, javascript: URIs, or inline event handlers such as onerror= and onload=
- Outbound browser requests from administrator sessions to unexpected external domains following a carbon console page load
- Unexpected DOM modifications or redirects reported by console users during routine administrative tasks
Detection Strategies
- Inspect stored XML payloads processed by the carbon console for HTML control characters and script-like tokens that survived input handling
- Review web server and reverse-proxy logs for POST requests to carbon management endpoints carrying XML bodies with embedded markup
- Correlate administrator browser telemetry with console access logs to identify script execution tied to specific stored records
Monitoring Recommendations
- Enable verbose audit logging on the carbon console for configuration changes and XML uploads
- Forward console access logs and browser-side Content Security Policy (CSP) violation reports to a central analytics pipeline for review
- Alert on first-seen external domains contacted by authenticated administrator sessions during console use
How to Mitigate CVE-2025-6623
Immediate Actions Required
- Apply the vendor fix referenced in WSO2 Security Advisory WSO2-2025-4316 as soon as it is available for your product version
- Restrict carbon console access to trusted administrative networks and remove any adjacent-network exposure that is not required
- Audit existing stored XML artifacts for script payloads and remove or sanitize affected entries
- Rotate high-privilege console credentials if unauthorized XML submissions are observed
Patch Information
WSO2 tracks the fix in the carbon-analytics-common pull request #870, which adds output encoding for the affected XML-rendering path. Consult WSO2 Security Advisory WSO2-2025-4316 for the authoritative list of fixed product versions and update instructions.
Workarounds
- Enforce a strict Content Security Policy on the carbon console to block inline scripts and unapproved script sources
- Place the console behind a web application firewall (WAF) with rules that reject XML bodies containing HTML or script constructs
- Limit console accounts with high privileges to the smallest viable group and require multi-factor authentication for those accounts
# Example CSP header for the carbon console reverse proxy
add_header Content-Security-Policy "default-src 'self'; script-src 'self'; object-src 'none'; frame-ancestors 'none'; base-uri 'self'";
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

