Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessCraft Special:Ask link with malicious separator
-
DeliveryDeliver crafted link to wiki user
-
ExploitVictim browser requests raw/special_page output
-
ExecutionUnescaped separator renders injected script in wiki origin
-
ImpactExfiltrate session cookie and CSRF token
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A05:2025 - Injection
ASVS 5.0
V1 - Encoding and Sanitization Output encoding, context-aware escaping
MITRE ATT&CK
2 techniques across 2 tactics: Initial Access, Execution The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | Exploitation requires the public Special:Ask special page with the remote request 'send-response' feature, which is enabled by default (extension.json:1765), so default installations are affected. The attacker must supply a crafted separator parameter (sep, propsep, valuesep, delim, or separator) together with request_type=raw for the list and category printers or request_type=special_page for the DSV and CSV printers. The list-printer path additionally needs at least two pages carrying semantic properties to satisfy the count>0 gate. The primary limiting factor is required victim interaction (UI:P): a user must click the attacker-supplied link, and the practical blast radius depends on that user's privilege level. No attacker authentication is required (PR:N). The interim mitigation per the advisory is removing 'send-response' from RemoteReqFeatures, which disables legitimate remote query functionality; the vendor fix is upgrading to the patched release. |
| Impact | This is a genuine but moderate-priority issue, not a drop-everything emergency. The CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:N/UI:P/VC:L/VI:L/VA:N, score 5.3 Medium) describes a reflected XSS that is network-reachable and low-complexity against default installations, but exploitation is gated on victim interaction (UI:P) - the attacker must persuade a user to click a crafted link, so it is not self-propagating or automatable. Impact is limited (VC:L/VI:L): script execution in the wiki origin enabling session cookie and CSRF-token theft, with the real-world severity scaling to the privilege of whoever clicks (administrative actions if an admin is lured). No EPSS score and no KEV listing were provided, so there is no confirmed active-exploitation signal; the vuln.today Decision label is 'this_month', consistent with a medium-severity, interaction-dependent web flaw rather than act_now/emergency. A detailed public PoC (exact request and expected response) is published in the GHSA advisory, which raises practical exploitability above a purely theoretical finding. Signals are broadly consistent (medium CVSS, no KEV/EPSS, this_month priority) with one data conflict noted below regarding fix status. |
| Exploit ScenarioDerived | An attacker crafts a Special:Ask URL with a malicious separator parameter (e.g. p[sep]=<img src=x onerror=...>) using request_type=raw and sends it to a wiki user via email or chat. When the victim - ideally a logged-in administrator - clicks the link, the browser receives the unescaped response served as text/html and executes the injected JavaScript in the wiki's origin, stealing the session cookie or CSRF token. A full working PoC request and expected response are published in the GHSA advisory, so the trigger is concretely documented rather than inferred. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 28, 2026
Reflected cross-site scripting in Semantic MediaWiki 3.0.0 through 7.3.0 lets remote attackers execute JavaScript in a wiki's origin by luring a victim to a crafted Special:Ask link that carries an unescaped separator parameter (sep, propsep, valuesep, delim, or separator) alongside request_type=raw or request_type=special_page. No attacker authentication is required, but the victim must click the link, so the practical blast radius depends on whether an administrator or other privileged user is targeted; successful script execution can read session cookies, exfiltrate CSRF tokens and perform actions as the victim. The vulnerable remote request path is enabled by default, and the flaw is a bypass of the earlier table-printer fix (GHSA-7xv3-gf2g-498h), which hardened only TableResultPrinter while six equivalent sinks in the list, category, DSV and CSV printers were left unescaped; the vendor has released 7.3.1, and a proof-of-concept request is published in the GitHub advisory. This is a genuine but moderate-priority issue for this month's patch cycle rather than a drop-everything emergency, and no confirmed active exploitation has been reported.
RemediationAI
Upgrade Semantic MediaWiki to the vendor-released fix version 7.3.1, which applies htmlspecialchars() output encoding (ENT_QUOTES | ENT_SUBSTITUTE, UTF-8) to the separator parameters in the list, category, DSV and CSV printers, following the pattern already used in TableResultPrinter::getValueSeparator(); installations pinned to an older branch should treat 7.3.1 as the minimum patched release rather than backporting piecemeal. If an immediate upgrade is not possible, the interim mitigation described in the advisory is to remove 'send-response' from the RemoteReqFeatures configuration (smwgRemoteReqFeatures), which closes the Special:Ask remote request path that serves the unescaped output - the trade-off is that legitimate remote query functionality, including external consumers of Special:Ask raw and file output, stops working. There is no separate Content-Type hardening that fully substitutes for the upgrade, since the encoding flaw sits in the printers themselves. Confirm the change on a staging wiki first, then validate that Special:Ask still returns expected results, and refer to the GitHub security advisory at https://github.com/SemanticMediaWiki/SemanticMediaWiki/security/advisories/GHSA-3v2c-r7rx-frc8 for the authoritative fix detail.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Semantic MediaWiki is a MediaWiki extension (Composer package mediawiki/semantic-media-wiki) that adds semantic annotation and a Special:Ask query interface. The root cause class is CWE-79, improper neutralization of input during web page generation: the separator parameters accepted by non-table result printers are joined into the response with PHP's implode() and receive no output encoding. The affected sinks in the 7.3.0 release are ListResultBuilder::getResultText (sep), SimpleRowBuilder::getRowText (propsep), ValueTextsBuilder::getValuesText (valuesep), CategoryResultPrinter (delim), DsvResultPrinter::getDSVLine (separator) and CsvFileExportPrinter::getCsv (separator). Two output paths matter: list and category printers (isExportFormat() = false) are dispatched through SMW_OUTPUT_RAW when request_type=raw, while DSV and CSV printers (isExportFormat() = true) go through SMW_OUTPUT_FILE when request_type=special_page, which also bypasses outputAsFile() and therefore the Content-Type header. Because parser sanitization in ResultPrinter::handleNonFileResult() only runs for SMW_OUTPUT_HTML - the recursiveTagParse branch requires that output mode - raw responses with no explicit Content-Type are interpreted by browsers as text/html and the injected markup is rendered. The earlier fix for GHSA-7xv3-gf2g-498h added htmlspecialchars() in TableResultPrinter::getValueSeparator() and was verified effective for the table printer in 7.3.0, but the same raw implode pattern in the six sibling sinks was never addressed; the exact version in which the request_type attack surface was introduced could not be established from the repository history, though the printers themselves date back to earlier releases. The trigger for each individual sink is a request-supplied separator value embedded verbatim in the response body, as detailed in the advisory's public proof-of-concept request.
Know when Semanticmediawiki gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:P/VC:L/VI:L/VA:N/SC:L/SI:L/SA:N
Network-reachable reflected XSS, no privileges needed (PR:N), but requires victim clicking a crafted link (UI:R); script runs in the victim's browser context (S:C) with limited session/token confidentiality and integrity impact and no availability loss.
Primary rating from Vendor (VulnCheck).
CVSS VectorVendor: VulnCheck
Lifecycle Timeline
1DescriptionCVE.org
SemanticMediaWiki from 3.0.0 before 7.3.1 contains a reflected cross-site scripting vulnerability in Special:Ask, which outputs separator parameters (sep, propsep, valuesep, delim, separator) without HTML escaping in list, category, DSV, and CSV result printers. Unauthenticated attackers can send victims crafted Special:Ask links using request_type=raw or request_type=special_page to execute JavaScript in the wiki's origin, stealing sessions or CSRF tokens.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Cross-Site Scripting occurs when an application accepts untrusted data and sends it to a web browser without proper validation or encoding.
Cross-Site Request Forgery exploits the automatic credential inclusion behavior of web browsers.
More in Semanticmediawiki
View allStored cross-site scripting in Semantic MediaWiki 3.1.0 through 6.x lets users with the MediaWiki 'edit' right inject ar
Semantic MediaWiki before 7.3.1 exposes Special:PropertyLabelSimilarity without an administrator restriction, and a call
Semantic MediaWiki 2.5.0 through 7.3.0 lets an editor with ordinary page-edit permission exhaust server CPU, memory and
Quadratic CPU consumption in Semantic MediaWiki's term parser gives unauthenticated visitors a way to tie up PHP search
Unauthenticated CPU and memory exhaustion in Semantic MediaWiki 3.0.0 up to (but not including) 7.3.1 allows remote atta
Cross-site request forgery in Semantic MediaWiki's Special:SMWAdmin page allows an attacker to drive destructive mainten
Semantic MediaWiki 2.5.0 through 7.3.1-alpha applies annotation-driven page protection without any authorization check,
Missing authorization in the Semantic MediaWiki smwtask API module allows unauthenticated remote visitors to invoke main
Semantic MediaWiki 7.3.0 through 7.3.1-alpha fails to authorize the target page in its smwtask update task, letting a ca
Missing authorization in Semantic MediaWiki 7.3.0 lets unauthenticated remote callers run queued maintenance jobs of arb
Reflected cross-site scripting in Semantic MediaWiki 7.0.0 through 7.1.0 lets an unauthenticated attacker execute arbitr
Reflected cross-site scripting in Semantic MediaWiki 7.1.0 and earlier lets an unauthenticated remote attacker inject ar
EUVD-2026-110703
GHSA-wqq7-744r-9mrr
NVD
CVE.org
EPSS
Exploit-DB