Unauthenticated read access to CI/CD job traces in GitLab CE/EE can expose sensitive CI/CD variable values because the GraphQL API fails to enforce proper authorization for trace retrieval, affecting all versions from 15.11 before 19.2.7, 19.3 before 19.3.3, and 19.4 before 19.4.1. Publicly available exploit code exists, with details disclosed through HackerOne report 3590929, and the bug was reported by GitLab itself; no confirmed active exploitation (CISA KEV) has been recorded. The CVSS score is only 3.7 (CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:N/A:N), reflecting high attack complexity and a confidentiality-only impact limited to reading trace contents, so this is a real but modest issue that manifests only under specific, non-universal conditions rather than an urgent, universally exploitable flaw.
OS command injection in RaspAP's raspap-webgui (raspap-webgui through version 3.5.5) lets an attacker who already holds a low-privileged authenticated session manipulate the cfg_id parameter of ajax/openvpn/del_ovpncfg.php to inject extra arguments/commands through the incomplete escapeshellcmd() filter. The flaw is remote but requires a valid session (CVSS 4.0 AV:N/AC:L/PR:L/UI:N, assessed as CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L), and because escapeshellcmd only neutralizes shell metacharacters rather than argument boundaries, the realistic outcome is constrained argument injection affecting integrity and availability rather than unrestricted remote code execution. Publicly available exploit code exists (a referenced GitHub gist), but there is no indication of confirmed active exploitation (CISA KEV); the vendor was contacted and did not respond, and no vendor-released patch has been identified at time of analysis.
OS command injection in RaspAP's raspap-webgui component (versions up to and including 3.5.5) lets an authenticated, high-privilege RaspAP administrator execute arbitrary operating-system commands on the host by manipulating the SSID value that WiFiManager::writeWpaSupplicant writes into the wpa_supplicant configuration. Exploit code has been published publicly and may be used, so the technical details are already circulating, although there is no confirmation of active exploitation (the CVE is not in CISA KEV). Impact is web-admin-to-host-OS escalation rather than external compromise, since valid administrative web GUI credentials are a hard prerequisite; the vendor was contacted early about the disclosure but did not respond, and no vendor-released patched version has been identified.
Command injection in MODSetter SurfSense through version 2.0.3 allows an authenticated low-privilege user to execute arbitrary shell commands on the host by sending a crafted payload to the /api/search-source/connectors/mcp/test endpoint within the MCP Connector Integration component. Exploitation is remote with low complexity and no user interaction, but requires valid credentials and a reachable/configured MCP connector feature. Publicly available exploit code exists (a proof-of-concept was published to a public gist and referenced by VulDB), and the vendor was contacted about the disclosure but did not respond; no official patched version is confirmed, and the signals on severity conflict (vendor CVSS 4.0 base 5.3 versus our assessed CVSS 3.1 vector of 8.8).
Cross-site scripting in serialize-javascript 7.1.1 arises because function values serialized by the library are not fully protected against script-closing tags: the SCRIPT_CLOSE_REGEXP escaping pattern can match across and swallow a second closing tag, so a surviving </script> terminates the embedding script element early and the remaining output is parsed as HTML in the page origin. Exploitation requires an application on the narrow vulnerable release to serialize a function value whose source is attacker-influenced and embed that output inside an HTML <script> element, and a victim must load the resulting page (user interaction); ordinary data values and options.isJSON output are unaffected. The flaw is fixed in 7.1.2, and no public exploit was identified at time of analysis.
Reflected cross-site scripting in the users.php script of coolbeans1212's MateisHomePage-Website lets remote, unauthenticated attackers inject arbitrary JavaScript into a page rendered to a victim who opens a crafted link. A successful attack requires the victim to click an attacker-supplied URL containing a payload in the 'search' GET parameter (CVSS:4.0 UI:P), so the flaw is non-persistent and must be delivered per victim rather than planted once. Publicly available exploit code exists and the issue has been publicly disclosed alongside a vendor commit, but there is no evidence of confirmed active exploitation (no CISA KEV listing); the independently assessed risk is LOW despite the network attack vector, because the product is a niche personal website and impact is limited to script execution in the victim's browser session (C:L/I:L, S:C).
Server-side request forgery (SSRF) in MODSetter SurfSense 0.0.36 and earlier allows an authenticated, low-privilege user to make the backend dereference attacker-influenced URLs by supplying crafted content to the Document Export Feature (the export_document route in surfsense_backend/app/routes/editor_routes.py). Because the server-side export/conversion pipeline (pypandoc/typst) fetches or processes references embedded in document content, a malicious export request can cause the server to issue outbound requests to resources it can reach. Publicly available exploit code exists and the flaw is remotely reachable, but the assessed impact is confidentiality-limited (VC:L) with no integrity or availability effect, making this a genuinely low-severity issue despite remote reachability.
Stored cross-site scripting in mall4j through 4.0 arises from unrestricted file uploads in admin FileController endpoints that require only any valid authenticated token and do not validate uploaded file types. An authenticated attacker can upload HTML or SVG files that execute script in an administrator's browser when the administrator opens the uploaded file from the local storage path, and publicly available exploit code exists. Real-world risk is modest because exploitation requires a token and privileged-user interaction, and the impact is client-side stored XSS rather than server-side code execution.
Origin validation error in OpenDMARC through version 1.4.2 lets remote attackers trigger incorrect organizational-domain extraction by sending mail from a specially crafted domain, potentially undermining DMARC alignment and allowing spoofed messages to evade DMARC-based filtering. The flaw carries a CVSS 4.0 base score of 5.3 and affects integrity only; publicly available exploit code exists, but there is no confirmation of active exploitation, and the vendor did not respond to disclosure so no official patch is available.
Stored cross-site scripting in CodeCanyon Rocket LMS up to and including version 2.2 lets an authenticated low-privilege user - for example a self-registered student - inject script through the Student Profile Image Upload component, which then executes in the browser of any admin, instructor, or fellow student who later views that profile; publicly available exploit code exists. Exploitation is not self-contained: it requires victim interaction (a second user must open the crafted profile) and results only in client-side script execution in the viewer's session, not server-side access. Because the vendor was contacted early but did not respond, no vendor-released patch is identified and every affected deployment remains exposed until compensating controls are applied.
Configuration backups from Netcore NAP930 firmware 0.1.241010.141410 can be decrypted by an attacker who obtains a backup artifact because the Backup/Restore component uses a hard-coded AES key (the aes_pass value in /lib/functions/backup_common.sh). The issue is remotely reachable but requires high administrative privileges (PR:H) to generate or retrieve a backup, and it only affects confidentiality, exposing backup contents rather than enabling writes or code execution; publicly available exploit code exists, and the vendor did not respond to early disclosure. This is a low-to-moderate risk finding for affected deployments, scoped to the specific NAP930 firmware build.
Server-side template injection in the Email Template component of ThinkCMF up to 8.0.7 lets an authenticated administrator submit template content that the underlying template engine evaluates server-side, enabling data exposure or logic execution within the application's privileges. Exploitation requires a privileged session able to reach MailController::templatePut in cmf-api/src/admin/controller/MailController.php, so anonymous and low-privilege users cannot trigger it; publicly available exploit code exists (a GitHub write-up targeting ThinkCMF v8.0.1), but there is no indication of confirmed active exploitation in CISA KEV. No vendor patch has been issued because the vendor did not respond to the disclosure, and independent assessment rates this a genuinely low-priority finding despite the SSTI classification.
A signed integer overflow (CWE-190) in Poppler's TrueType font conversion path, specifically FoFiTrueType::cvtSfnts in fofi/FoFiTrueType.cc, lets an attacker who supplies a crafted sfnt/TrueType font - typically embedded inside a PDF that is rendered or converted with tools such as pdftops - corrupt the handling of font table lengths during conversion. The defect affects Poppler 26.06.0, 26.07.0 and 26.08.0 and is reachable only locally, when a victim or local process opens or converts the malicious file; the assessed impact is bounded to limited integrity loss (VI:L) with no confidentiality or availability impact, so this is not a straightforward remote code execution issue. Publicly available exploit code exists and has been disclosed alongside a signed-overflow write-up in a public GitHub repository, and upstream has published a fix as commit 245d3c6823377755f2c1d5fdddd010279c6ed94d (merge request !2326); despite the public PoC, the CVSS v4.0 score of 4.8 and the vendor vector (AV:L/PR:L, VI:L, E:P) keep this a low-priority issue for most deployments.
Signed integer overflow in the Splash rasterization backend of Freedesktop Poppler (SplashClip::clipToPath in splash/SplashClip.cc) lets a local attacker crash or destabilize any Poppler-backed application that renders an attacker-crafted PDF, yielding a denial-of-service primitive with no confidentiality or integrity impact. Versions up to and including 26.08.0 are affected, publicly available exploit code exists (no KEV listing, so this is not confirmed active exploitation), and the flaw is fixed in 26.09.0. Exploitation is local in nature and requires the victim to open or render the malicious document, so despite the alarming CWE-190 integer-overflow class the realistic risk is low and availability-only.
Authenticated attackers positioned on the adjacent network can abuse unsafe inline-script and string-evaluation (eval-class) functions in Toptech TMS7 and Toptech TopHat terminal-management deployments, achieving limited code injection with Low confidentiality and integrity impact and no availability effect. The issue is scored 2.1 under CVSS 4.0 and was independently assessed as AV:A/AC:H/PR:L/UI:N with Low C/I, so it sits firmly in the low-priority tier despite its CWE-95 code-injection classification: it is not reachable from the internet by default, requires valid low-privilege credentials, and demands high attack complexity to land the crafted input on the evaluator. No public exploit code has been identified at time of analysis, the issue does not appear on CISA KEV, and no EPSS score was supplied in the source data, so exploitation is not confirmed to be active.
QUIC-enabled servers running OpenSSL 3.5.0 through 3.5.8, 3.6.0 through 3.6.4, and 4.0.0 through 4.0.2 can be induced to over-account the amplification credit granted to an unvalidated client address, so the library may emit more reflected UDP traffic toward a spoofed source than QUIC's anti-amplification rules permit. The flaw sits in the pre-address-validation path of the QUIC stack, meaning unauthenticated remote attackers who can spoof source IPs need no credentials or user interaction, though the assessed CVSS vector (AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:L) and score of 3.7 reflect the narrow timing and traffic conditions required and a bounded impact limited to traffic amplification rather than code execution or data exposure. No public exploit code is identified at time of analysis, it is not confirmed actively exploited (no CISA KEV listing), and CISA's SSVC assessment records exploitation as none, automatable as no, and technical impact as partial.
OpenSSL's SM2 scalar multiplication on ARM64 and RISC-V is not constant-time, allowing a remote, unauthenticated attacker (CVSS PR:N) who can precisely measure many SM2 operations to conduct statistical timing analysis and potentially recover key material. The flaw affects OpenSSL 4.0.0 before 4.0.3, 3.4.0 before 3.4.8, 3.5.0 before 3.5.9, and 3.6.0 before 3.6.5; practical exploitation is sharply constrained because the target must use SM2 on an affected architecture and the attacker needs high measurement volume and precision. No public exploit code or CISA KEV entry was identified at time of analysis, and vendor patches are available; the 'High' severity of the OpenSSL 4.0.3 release pertains to a sibling CVE, not this low-rated issue (CVSS 3.7).
A timing side-channel in OpenSSL's elliptic-curve scalar multiplication for non-NIST (non-standard or custom) curves can leak information about private-key material to an attacker who can induce and precisely measure many such operations. The affected branches are OpenSSL 1.0.2 before 1.0.2zs, 1.1.1 before 1.1.1zj, 3.0 before 3.0.23, 3.4 before 3.4.8, 3.5 before 3.5.9, 3.6 before 3.6.5, and 4.0 before 4.0.3; deployments that use only the standard NIST curves (P-256/P-384/P-521) are not exposed. Assessment gives it low severity (CVSS 3.7, AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:N/A:N, CWE-208): no public exploit code was identified at time of analysis, no confirmed active exploitation (not in CISA KEV), and SSVC rates exploitation as none and automatable as no, with the high-precision timing recovery and the specific curve configuration sharply limiting realistic abuse.
A timing side-channel in SM2 signature generation in OpenSSL lets a remote, unauthenticated attacker who can precisely measure how long signing operations take to infer information about the SM2 private key, producing limited confidentiality impact (CVSS 3.1 base score 3.7, vector AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:N/A:N). Affected branches are 1.1.1 before 1.1.1zj, 3.0.0 before 3.0.23, 3.4.0 before 3.4.8, 3.5.0 before 3.5.9, 3.6.0 before 3.6.5, and 4.0.0 before 4.0.3, with the fix published in the OpenSSL 4.0.3 security release. No public exploit code has been identified at time of analysis, CISA SSVC records no exploitation activity (Exploitation: none, Automatable: no, Technical Impact: partial), and real-world exposure is sharply narrowed to deployments that actually perform SM2 signatures and are reachable over a low-jitter network path.
Russh SSH servers running versions prior to 0.62.6 that set server::Config.max_auth_attempts to a nonzero value silently fail to enforce that limit, letting an unauthenticated remote client keep sending USERAUTH_REQUEST messages on a single connection past the configured cap. The auth_attempts counter is incremented on the server::run_stream path in russh/src/server/encrypted.rs but is never compared against the configured maximum, so the intended 'too many authentication attempts' disconnect never fires, preserving unlimited online password guessing and adding backend authentication load. Practical impact is limited: the attacker must still guess valid credentials, the bug has no effect when max_auth_attempts is 0 or when external throttling (fail2ban, firewall rate limits) is used, and no public exploit code has been identified at time of analysis.
Russh, the Rust SSH client and server library, can be driven into a panic-driven connection teardown by a remote peer whenever a deployment explicitly permits mac=none in its algorithm preferences. All releases prior to 0.63.1 are affected, because negotiation validated the cipher's needs_mac() requirement only on the MAC-selection-failure branch, allowing a MAC-requiring CTR or CBC cipher such as aes256-ctr to be paired with a zero-length MAC; the attacker then sends a packet with a decrypted length of zero, shrinking the buffer read in russh/src/cipher/mod.rs and panicking the indexing at buffer.buffer[16..]. The resulting impact is availability-only (the connection task terminates) with no code execution or data exposure, requires the non-default mac=none configuration and is rated high-complexity per the assessed AV:N/AC:H/PR:N/UI:N vector (unauthenticated remote attackers per the CVSS vector), and no public exploit code has been identified at time of analysis.
Malformed UTF-8 in a filename can bypass the control-character check performed by Flysystem's default WhitespacePathNormalizer before version 3.35.3, letting a filename containing ANSI escape sequences survive normalization and be stored via Filesystem::write(). When an administrator later renders Filesystem::listContents() output in an ANSI-interpreting terminal, those sequences can hide or spoof entries in the listing, giving a malicious write-capable user (PR:L) a terminal-output deception primitive with UI:R. The issue is confined to spoofed or hidden directory listings - there is no code execution, no data disclosure, and no integrity impact beyond the stored filename itself; no public exploit code has been identified at time of analysis and there is no confirmed active exploitation, and the CVSS 3.5 score reflects the low overall risk.
Information disclosure in Google Chrome on Android before version 154.0.8037.92 allows an attacker who has already compromised the renderer process to read memory outside the browser sandbox by getting the victim to load a crafted HTML page. The flaw is an uninitialized-resource bug (CWE-908) in the GPU component, so it leaks memory rather than executing code or modifying state, and the practical exploitability is sharply limited by the precondition that the attacker must first hold a compromised renderer - a requirement the scored vector does not capture despite PR:N. Scored CVSS is only 3.4 (AV:N/AC:H/PR:N/UI:R/S:C/C:L/I:N/A:N), no public exploit code was identified at time of analysis, and no confirmed active exploitation (CISA KEV) is associated with this CVE; a vendor-released fix exists in Chrome 154.0.8037.92, making this a low-priority patch-and-move-on item despite Chromium's 'High' security severity label.
Memory disclosure across the Chrome sandbox boundary affects Google Chrome on Windows prior to 154.0.8037.92, where an uninitialized resource in the Media component lets content read memory it should not be able to reach. Exploitation is not self-contained: the attacker must already have compromised the renderer process through a separate, earlier vulnerability, and then have the victim load crafted HTML that exercises the Media code path (CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:C/C:L/I:N/A:N, base 3.4). Impact is limited to low-confidentiality readout with no integrity or availability effect, making this realistically an exploit-chain building block, and there is no public exploit identified at time of analysis; a vendor patch is available in 154.0.8037.92.
An uninitialized-memory flaw in Chrome's GPU component (CVE-2026-102319, CWE-908) lets an attacker who has already compromised the renderer process read memory outside the browser sandbox by loading a crafted HTML page, providing a sandbox-weakening information-disclosure primitive rather than direct code execution. Affected are Google Chrome builds prior to 154.0.8037.92 on all platforms (CPE cpe:2.3:a:google:chrome:*:*:*:*:*:*:*:*); the independent assessment scores this CVSS 3.1 at 3.4 (AV:N/AC:H/PR:N/UI:R/S:C/C:L/I:N/A:N), classifying it as a genuine but low-priority issue despite Chromium's internal 'High' label, because it requires a prior renderer compromise, high attack complexity, and user interaction, and yields only partial confidentiality impact. No public exploit code and no confirmed active exploitation (CISA KEV) were identified at time of analysis, and vendor-released patch Chrome 154.0.8037.92 is available.
A local buffer overflow in the underlying operating system of HPE Networking Instant On devices (all builds up to and including 3.4.1.0) allows a low-privilege authenticated local attacker to interrupt the normal operation of a single affected service. The flaw is rated CVSS 3.1 3.3 (AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L), meaning it is not remotely reachable, requires an existing low-privileged local foothold on the device OS, and is limited to partial availability impact on one service with no confidentiality or integrity loss; no public exploit identified at time of analysis, and no CISA KEV listing or EPSS scoring is associated with this CVE. Overall risk is low and the pre-existing authenticated local access requirement is the dominant limiting factor.
Authenticated path traversal in the HPE Networking Instant On command line interface allows an attacker who already holds administrative CLI access to steer file operations outside the intended directory, modifying a limited set of files on the underlying operating system and interrupting normal operation of the affected service. Only Instant On versions up to and including 3.4.1.0 are affected, and the vector is local (AV:L) with high attack complexity (AC:H) and high privileges required (PR:H), giving a CVSS 3.1 base score of 3.0. There is no data disclosure, no integrity or availability impact outside the limited file set and service interruption, no CISA KEV entry, and no public exploit identified at time of analysis.
Buffer overflow in the API endpoint of HPE Networking Instant On lets an authenticated attacker who already holds administrative privileges interrupt the affected service, producing a short-lived denial of service that recovers on its own without manual intervention. Only administrators can reach the vulnerable code path (CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:L), and there is no confidentiality or integrity impact, so the practical exposure is limited to brief, self-healing service interruptions. All Instant On versions up to and including 3.4.1.0 are reported affected; no public exploit code has been identified at time of analysis and the issue is not listed as confirmed actively exploited (CISA KEV), which is consistent with the low 2.7 base score.
Low-privileged pretix team members who can reach a specific event but who were deliberately denied access to that event's orders can read two aggregated business figures they should not see: the event's attendee count and its total revenue. Exploitation requires an authenticated account already placed on the organizer's team with event-level access but without order-level permission, so there is no path for an outside, unauthenticated attacker, and the disclosure is confined to aggregate numbers rather than individual order or personal data. No public exploit code has been identified at time of analysis, and the issue is assessed as a genuinely low-risk, low-priority information-disclosure bug affecting pretix installations before 2026.7.1.
Cache poisoning in UnoPim before 2.0.1 and 2.1.x before 2.1.1 allows unauthenticated remote attackers to inject an arbitrary origin into admin layout pages by sending a crafted X-Forwarded-Host header, because the application trusts every connecting client as a proxy and never validates the header. When a shared caching proxy or CDN stores the poisoned admin response, a subsequent administrator loading that page has the admin UI's JavaScript assets resolved from the attacker's server, executing attacker-supplied code inside an authenticated session. Exploitation is conditional - it requires an intermediary cache in front of admin responses and an administrator visiting the poisoned page (assessed CVSS 3.1 AV:N/AC:H/PR:N/UI:R/S:C, with limited confidentiality and integrity impact), no public exploit code or confirmed active exploitation was identified at time of analysis, and the CVSS 4.0 base score is 2.3.
API-uploaded files in pretix are not correctly bound to the uploader's authentication method, and a single session token is shared across all token-based API users, undermining the intended isolation. This affects pretix versions prior to 2026.7.1. Exploitation requires a valid token-based API account and knowledge of the randomly generated UUID of a target file, which only exists for about a day, making real-world risk very low (CVSS 2.3). No public exploit code has been identified at time of analysis, and the issue is not known to be actively exploited.
Server-side request forgery in Fider before 0.38.0 lets an authenticated administrator who controls DNS for a configured hostname redirect the server's outbound HTTP requests to internal services or cloud metadata endpoints by exploiting a time-of-check/time-of-use gap between URL validation and the actual connection. The exposure is narrow: only webhook URLs and custom OAuth provider token/profile URLs are user-configurable outbound targets, the attacker needs admin privileges and must win a DNS-rebinding race (high attack complexity), and instances with AllowPrivateNetworkTargets deliberately enabled are out of scope, so this remains a low-priority issue despite the network vector (CWE-918). No public exploit code has been identified at time of analysis and there is no evidence of active exploitation; v0.38.0 fixes the flaw by re-checking the resolved IP at dial time.
OAuth access tokens issued by pretix continue to be honored for their entire original lifetime even after an administrator manually disables the OAuth client (application) they belong to, so disabling an application is not an effective immediate revocation mechanism. Exploitation requires an attacker to already possess a valid, unexpired access token that was issued to that client before it was disabled, meaning this is an authenticated, high-complexity access-control weakness rather than a remotely reachable bypass. Severity is low (vendor CVSS 4.0 score 2.1; independent CVSS 3.1 assessment 2.1 with C:L/I:L, AC:H, PR:L), no CISA KEV listing is present and no public exploit code has been identified at time of analysis, and no EPSS score was supplied in the available intelligence. Because the exposure window is bounded by the token's original TTL and cannot be extended, the practical risk is limited to the residual lifetime of already-issued tokens.
Stored HTML injection in pretix lets an authenticated user with organizer permissions plant malicious markup inside the help texts of configurable fields, which is then rendered to other users who view the affected pages. Exploitation is gated by the organizer privilege level, requires a victim to subsequently load the injected content (UI:P), and the vendor rates it high-complexity with an additional attack requirement (AC:H/AT:P), so it is not reliably repeatable against arbitrary targets. No public exploit code and no confirmed active exploitation were identified at time of analysis, and the CVSS 4.0 base score is only 2.1, reflecting limited confidentiality and integrity impact with no availability loss.
Object-level authorization bypass in owen2345 Camaleon CMS through 2.9.2 lets an authenticated user who holds only the media-library permission (`:manage, :media`) overwrite another user's avatar by calling the admin media crop endpoint with the `saved_avatar` parameter set to a victim's user ID. The crop action reaches the same avatar meta write that `UsersController#update` protects with `:manage, :users`, but enforces only the media permission, so the media library becomes a side door into profile editing; post-PR-#1267 the target is scoped to `current_site.users`, limiting the bypass to same-site, cross-user writes rather than cross-tenant ones. Publicly available exploit code exists (CVSS 4.0 vector carries E:P), and the vendor has released a fix in 2.9.3; the exposure is limited to profile-meta tampering with no access to passwords, credentials, or confidential data, so the assessed risk is genuinely low. No confirmed active exploitation (CISA KEV) was identified at time of analysis.
Unrestricted file upload in mwasikz robo-cafe-rms (the manage-food.php food-management endpoint) lets an attacker with an authenticated high-privilege admin session upload arbitrary files to the web server over the network, which can lead to remote code execution if the server executes uploaded PHP from the upload directory. All deployments are affected, since the project ships as a rolling release with no versioned affected or fixed builds (CPE cpe:2.3:a:mwasikz:robo-cafe-rms:*:*:*:*:*:*:*:*). Publicly available exploit code exists, and the vendor was contacted early but did not respond, so no vendor-released patch has been identified; VulDB rates the issue CVSS 4.0 5.1 (Medium), while our independent CVSS 3.1 assessment of AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H places it higher (7.2 High) because successful upload can yield full confidentiality, integrity and availability impact on the host. The main limiting factor is the admin-authentication prerequisite (PR:H), which makes this an authenticated post-compromise escalation rather than an unauthenticated remote compromise.
Stored cross-site scripting in mwasikz robo-cafe-rms allows an authenticated low-privilege user to inject arbitrary script into the profile Name, Address, or City fields, which then executes in the browser of any victim who views the affected profile page. Publicly available exploit code exists, and because the project follows a rolling-release model, no version boundary or vendor patch can be cited for the affected or fixed code. The risk is low-to-moderate in practice: exploitation requires a valid account, the injected content only fires when another user loads the poisoned view, and the product has a very small deployment base. Vendor contact was attempted early in the disclosure process and produced no response.
Session fixation in pretix allows an attacker to trick a victim into following a crafted link that silently logs the victim into the attacker's own customer account, so order details the victim subsequently enters can be captured into that attacker-controlled account. The behavior only occurs in a non-default cross-domain deployment where the event is served from a different domain than the organizer page, and it additionally requires victim interaction and the victim failing to notice the unfamiliar logged-in account before completing an order, which keeps practical reliability low; the issue is assessed as genuinely low risk (CVSS 3.1 AV:N/AC:H/PR:N/UI:R/S:U/C:L/I:L/A:N). No public exploit code or active exploitation has been identified at time of analysis, and pretix has published a fix in its 2026-7-1 release.
Information disclosure in Intelbras TIP 125i VoIP desk phones running firmware 4.3.35 or 4.3.41 allows an attacker who already holds a low-privilege account on the device's web management interface to read sensitive data that the 'Básico' configuration page embeds directly in its page source. Exploitation is remote but rated high-complexity and 'difficult' by the vendor, requiring network reachability to the phone's administrative UI plus at least low-level privileges; there is no user-interaction requirement and no integrity or availability impact - only a limited confidentiality leak. Publicly available exploit code exists, but the flaw is not confirmed actively exploited, and the CVSS 4.0 base score of 1.3 reflects the constrained, low-value exposure.
Authenticated users of the pretix ticketing platform can bypass validation logic built into the checkout flow by skipping entire check-in steps, allowing data that those steps were meant to verify to pass through unchecked. The flaw requires a legitimate participant in the checkout process (PR:L) and only affects the integrity of check-in/checkout data - there is no confidentiality or availability impact and no victim interaction is required. Only deployments running pretix versions prior to the 2026.7.1 release are exposed, and no public exploit identified at time of analysis; the vendor's own scoring is low (CVSS 1.3), matching an independent assessment of low real-world risk despite the 'Authentication Bypass' tag.