Remote code execution in AnyDesk before 8.0.3 allows unauthenticated attackers to execute arbitrary code as root on Linux by sending a specially crafted packet to the session protocol's stream packet handler for mode 5. The victim does not need to approve desktop control, and publicly available exploit code exists. This is a critical heap-based buffer overflow (CWE-130) with a CVSS score of 9.8, though the assessed root-level impact is specifically demonstrated on Linux and other platforms are not confirmed.
Unauthenticated OS command injection in the Web Management Backend of TRENDnet TEW-751DR, TEW-752DRU, and TEW733GR routers (firmware 1.03B01/1.03B03) allows remote attackers to execute arbitrary shell commands on the device by sending crafted input in the 'service' argument handled by the fwrite function of pigwidgeon.cgi. Per the CVSS vector (PR:N/UI:N/AC:L) no authentication, user interaction, or non-default configuration is required; the principal limiting factor is reachability of the management interface, which is LAN-side by default and only exposed over WAN where remote administration is enabled or the router is internet-facing. Publicly available exploit code exists (a GitHub Proof-of-Concept targeting these models), so exposed units should be treated as compromisable with minimal attacker effort; EPSS and CISA KEV data were not provided, so confirmed active exploitation cannot be asserted. Vendor has confirmed the affected models reached end-of-life in 2011 and will not be remediated by a firmware fix.
Unauthenticated command injection in Edimax BR-6208AC firmware version 1.32 lets remote attackers execute arbitrary shell commands on the router by submitting crafted input to the tftpsIP parameter of the /goform/formAutoFWupgrade handler, which normally accepts a TFTP server address for firmware fetch operations. The vendor was contacted about the disclosure but did not respond, and the device is an end-of-life product no longer supported by the maintainer, so no vendor-released patch is expected; publicly available exploit code exists (a GitHub proof-of-concept under an IoT vulnerability repository) and the CVSS 4.0 score of 9.3 reflects a full confidentiality, integrity, and availability impact. The vulnerability is not listed in CISA KEV (no confirmed active exploitation reported), but pre-auth reachability of the HTTP management interface - particularly if WAN administration is enabled - makes this a high-priority exposure for anyone still running the device.
Improper authentication in the CGI handling of Netcore NR268 firmware 1.7.121109 allows remote, unauthenticated attackers to bypass authentication on the device's Boa-based web management interface via the user_ok function, which fails open rather than denying the request. Publicly available exploit code exists (a detailed report is published on GitHub), and the vendor was contacted but did not respond or release a fix. Exposure is the practical risk driver: devices whose HTTP management/CGI interface is reachable from the WAN or an untrusted LAN segment are at high risk, while deployments restricting management to a trusted internal interface are substantially lower risk.
A stack-based buffer overflow in the Mac Filter configuration handler of Linksys E1700 firmware 1.0.0.4.003 allows an attacker who can reach the router's administrative web interface to send an overlong newap_text_%d value to /goform/ssid1MACFilter and corrupt the stack, resulting in a crash and potentially code execution with the privileges of the device's web server. Publicly available exploit code exists (a detailed PoC is published on GitHub), but there is no CISA KEV listing and active exploitation has not been confirmed, so this should be treated as a credible but not yet observed attack. The vendor was contacted early about the disclosure and did not respond, so no patched firmware has been confirmed; the provided CVSS 4.0 vector rates it 9.3 with PR:N, while our independent assessment maps the same flaw to CVSS 3.1 AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H, because the /goform endpoint sits behind the admin UI and is normally reachable only after authenticating - resolving that pre-auth vs. post-auth question is the single biggest factor in real-world risk.
Unauthenticated remote code execution is achievable in Square Squash (all versions through commit e73f280) through a two-stage YAML deserialization path in the symbolication API. An attacker first stores crafted YAML under a chosen symbolication UUID using the unauthenticated POST /api/1.0/symbolication endpoint; the payload only detonates when occurrences referencing that UUID are later processed by the symbolication worker, at which point the stored value is passed to YAML.load with SafeYAML running in unsafe mode, permitting arbitrary Ruby object instantiation and code execution. Publicly available exploit code exists, and no CISA KEV listing confirms active exploitation; exploitation is deferred rather than immediate, which is why the assessed complexity is high (AC:H) despite requiring no authentication or user interaction.
Authenticated attackers with low privileges can escalate to elevated privileges on IBM i systems running Db2 Mirror for i 7.4, 7.5, or 7.6 by submitting a crafted request that bypasses additional authorization and input-validation checks, potentially allowing malicious code execution. This critical-severity (CVSS 9.9) privilege escalation affects standard deployments and requires a valid low-privilege account; no public exploit code or active exploitation has been identified at time of analysis, but a vendor patch is available.
Remote code execution and sensitive-information disclosure in IBM i 7.3, 7.4, 7.5, and 7.6 stem from improper data sanitization by the HTTP Server Administration web application (CWE-94 code injection). An authenticated attacker with only low privileges who can reach that administration interface over the network can inject and execute arbitrary code or read sensitive data, with the scope change (S:C) driving Confidentiality, Integrity, and Availability impacts all High for a CVSS 3.1 base score of 9.9. Exploitation is not unauthenticated and is not possible against systems that do not run the HTTP Server Administration interface, or that expose it only to a trusted management network; no public exploit has been identified at time of analysis and there is no CISA KEV confirmation of active exploitation, though IBM has published a vendor patch.
Authenticated path traversal in IBM DataStage on Cloud Pak for Data 5.4.0.0 allows a remote attacker holding a valid platform account to bypass intended directory restrictions and execute arbitrary code. Exploitation requires only low-privileged credentials (CVSS PR:L), needs no user interaction, and is network-reachable with low complexity, which is why the vector scores 9.9 Critical with a scope change (S:C) reflecting impact that can extend beyond the vulnerable component. Despite the high score, the practical risk picture is more moderate: EPSS is only 0.49% (40th percentile), CISA SSVC rates exploitation as none with automatable set to no, there is no public exploit code identified at time of analysis and no KEV listing, and the vendor has published a patch, so remediation is available rather than dependent on compensating controls alone.
Authenticated privilege escalation in Progress MarkLogic Server lets an attacker who already holds query-execution access invoke an XQuery evaluation function without the additional evaluation privilege that function requires, so the query is evaluated with elevated permissions and can lead to server-side code execution in the server's security context. Affected deployments are 12.x releases before 12.1.0 and 11.x releases before 11.3.7; the vulnerability is rated CVSS 9.9 (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H), with the scope change and full confidentiality, integrity, and availability impact reflecting what is possible once the missing privilege check is bypassed. Exploitation requires a pre-existing account with query-execution access rather than anonymous reachability, and no public exploit code has been identified at time of analysis; there is no confirmed active exploitation (CISA KEV) reported for this issue.
Authenticated XQuery injection in Progress MarkLogic Server 11.x before 11.3.7 and 12.x before 12.1.0 allows a remote attacker who already holds a manage-user-level account to escalate to persistent unauthorized administrative control of the database. The flaw resides in a deprecated Management API package path and carries a 9.9 CVSS score because exploitation crosses a security boundary (S:C) and fully compromises confidentiality, integrity, and availability. It is not a pre-authentication issue - a valid manage-user credential is mandatory - and no public exploit code has been identified at time of analysis, so risk is concentrated in environments where the Management API is network-exposed and manage-user grants are loosely administered.
An authenticated privilege-escalation flaw in Progress MarkLogic Server lets an attacker who already holds forest-management privileges create a persistent, unauthorized administrator account, resulting in complete administrative control of the server. Affected are 12.x releases before 12.1.0 and 11.x releases before 11.3.7; the vulnerability is reachable over the network (AV:N) with low privileges (PR:L) and no user interaction, and it is rated 9.9 because a successful escalation changes scope, giving the attacker high confidentiality, integrity and availability impact across the deployment. No public exploit code has been identified at time of analysis, and there is no confirmed active exploitation (no CISA KEV listing); the practical risk is concentrated in deployments that delegate forest-management rights to non-super-admin or duty-separated roles, since an attacker must first possess that privilege to exploit the flaw.
Improper authorization in obsidian-tc prior to 1.3.1 lets any holder of a valid JWT reach beyond its own vault and read, write, or delete the contents of every other configured vault on the same HTTP server. The flaw affects CWE-639-style trust in a caller-controlled key: tool dispatch never compares the caller-supplied input.vault value against the vault bound to the caller, the JWT carries no vault claim, and the folder ACL is a single global instance, so roughly 90 vault-scoped tools resolve an attacker-named vault id under that shared ACL. Exploitation requires the HTTP transport with more than one vault configured and a valid token (authenticated, PR:L); the trusted stdio transport is unaffected and single-vault deployments have no second vault to pivot to, but cross-tenant confidentiality, integrity, and availability are all fully compromised when those conditions hold - no public exploit code has been identified at time of analysis.
Remote code execution in IBM Maximo AI Service 9.1.0 and 9.2.0 lets an unauthenticated remote attacker execute arbitrary code on an affected deployment by abusing improper control of code generation (CWE-94). With a CVSS 3.1 base score of 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H), the flaw requires no authentication, no user interaction, and no special conditions beyond network reachability of the vulnerable instance. No public exploit code has been identified at time of analysis, and there is no KEV listing; IBM has published an advisory with a vendor fix, so the practical priority is identifying and patching internet- or network-exposed Maximo AI Service 9.1.0/9.2.0 instances.
A stack-based buffer overflow (CWE-121) in the UNISOC modem OTA stack lets an attacker within radio range achieve remote code execution or crash the baseband by broadcasting a crafted SIB1 message whose scheduling information is not properly validated, which the modem parses automatically with no authentication or user interaction (PR:N/UI:N). The vendor assigned a CVSS 3.1 score of 9.8 with an AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H vector, but the exploitation path is adjacency-gated: the adversary must operate a rogue cellular base station (SDR plus a full baseband stack), out-compete legitimate cells so the victim device camps on the malicious cell during idle-mode (re)selection, and the target must run the affected UNISOC modem/OTA stack - making this a proximity-limited, targeted attack rather than an internet-scale one. No public exploit code and no confirmed active exploitation were identified at time of analysis.
Unauthenticated OS command injection in GPT-SoVITS through 20250606v2pro lets any network-reachable attacker run arbitrary commands as the server user by inserting shell substitutions such as $(...) into the UVR5 WebUI output folder textbox and then triggering an mp3 or m4a export. The root cause (CWE-78) sits in tools/uvr5/vr.py, which hands the user-controlled output path to a shell during export without sanitization, so the impact is full compromise of the account running the WebUI (high confidentiality, integrity and availability loss per both the supplied CVSS 4.0 score of 9.3 and the independent CVSS 3.1 assessment of AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H). Exploitation depends on the UVR5 WebUI being reachable over the network, which is not guaranteed because these Gradio-style instances are commonly bound to localhost or a trusted LAN, and no public exploit code or confirmed in-the-wild activity has been identified from the available intelligence.
Authentication bypass and privilege escalation in IBM Planning Analytics 2.0 and 2.1 allow an unauthenticated remote attacker to reach functionality that should require credentials by abusing improperly validated sub-requests (CWE-288). Per the assessed vector CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H, exploitation needs only network reachability to the Planning Analytics application interface - no authentication, no user interaction, and no documented non-default configuration - yielding high confidentiality, integrity, and availability impact, which is why the score is 9.8. Signals conflict and must be weighed carefully: despite the severity and total technical impact under SSVC, EPSS is only 0.36% (28th percentile), CISA SSVC records Exploitation: none, and no public exploit code has been identified at time of analysis, so this should be treated as a serious but not yet observed threat. The vendor has released a fix, so exposure is concentrated in unpatched, internet- or partner-reachable deployments.
Unauthenticated remote attackers can gain full administrative control of the HAProxy Data Plane API - and through it the managed HAProxy load balancer - in Community and Enterprise Edition releases 2.2.0 through 3.4.3 when a specific mTLS topology is deployed. When a TLS Certificate Authority (tls_ca) and a numeric TLS port (tls_port) are configured, Server.ConfigureAPI() sets a process-global flag that disables HTTP Basic authentication on a request handler shared concurrently by HTTP, HTTPS, and UNIX-domain-socket listeners, while mutual-TLS client-certificate verification is only enforced on the HTTPS path; every other active listener therefore fails open. An attacker who can reach a plain HTTP or UNIX management endpoint can read cleartext HAProxy configuration data and replace the managed configuration via the raw configuration endpoint; no public exploit code has been identified at time of analysis, and the vendor has released fixed versions 3.4.4, 3.3.9, 3.2.17, 3.0.25 and 2.9.29 (plus EE backports).
Unauthenticated remote agent enumeration and invocation in PraisonAI's AgentTeam.launch() / Agents.launch() HTTP server lets any client that can reach the listening port list deployed agents and trigger agent.chat() calls, burning model resources, reaching connected private context, and firing agent tools and workflow side effects. The affected artifact is the Python package praisonaiagents before 1.6.59, whose FastAPI server binds to 0.0.0.0 by default and accepts requests with no Authorization header - or even with an obviously wrong bearer token. Vendor-released patch: praisonaiagents 1.6.59 (PraisonAI v4.6.59); no public exploit identified at time of analysis, and the practical limiting factor is deployment: the routes are only remotely reachable if launch() is actually started and network-exposed rather than bound to localhost or placed behind an authenticating proxy.
Loading an untrusted EXL3-format checkpoint in ExLlamaV3 can drive an out-of-range CUDA kernel function-pointer read and invocation, reliably crashing the process (denial of service) during model load. The EXL3 loader derives K from the model-supplied trellis.shape[-1] // 16 and computes kernel index cbi from it, then indexes 24-entry reconstruct_kernel_instances/reconstruct_had_kernel_instances tables without bounds checks in reconstruct_slice and reconstruct_had_slice; the authoritative assessment rates exploitation as local, high-complexity, user-interaction-required (AV:L/AC:H/PR:N/UI:R) with high availability impact, and controlled indirect kernel dispatch beyond the crash only under specific host memory layouts. No public exploit was identified at time of analysis, EPSS is low (0.20%, 10th percentile), and upstream fix PR #310 adds TORCH_CHECK bounds.
Unauthenticated command injection in Edimax BR-6208AC firmware version 1.32 allows a remote attacker to execute arbitrary operating-system commands on the router by supplying crafted input through the readE2P argument of the /goform/formWlanMP endpoint. The issue is rated critical (CVSS 4.0 base 9.3; an independent assessment reaches the equivalent CVSS 3.1 vector AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) and publicly available exploit code exists, although there is no confirmation of active exploitation (device is not in CISA KEV). The vendor was contacted early and did not respond, and the product is no longer supported, so the decisive practical concern is permanent exposure: any BR-6208AC 1.32 whose web management interface is reachable over the WAN or from an untrusted network segment remains exploitable indefinitely.
Denial of service in IBM DataStage on Cloud Pak for Data 5.4.0.0 can be triggered by a remote user who already holds valid platform credentials and supplies path traversal sequences to the product's file-handling logic, disrupting or halting the service. Exploitation requires an authenticated session with low privileges (CVSS vector PR:L) and no action by a victim, so the practical barrier is credential access rather than network reachability; availability is the realistic impact, with no evidence of code execution. A vendor patch is available via IBM support advisory node 7288649, and no public exploit code or confirmed active exploitation was identified at time of analysis.
Privilege abuse in Progress MarkLogic Server 12.x before 12.1.0 and 11.x before 11.3.7 lets a remote attacker who already holds delegated database-administrator privileges run backup or restore operations against forests outside their authorized database scope, resulting in unauthorized read or modification of security-critical data. The flaw is rated CVSS 9.6 (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:N) under CWE-269 (Improper Privilege Management), and because exploitation requires the pre-existing delegated DB-admin role it is an authenticated privilege-abuse issue rather than an unauthenticated pre-auth bug. No public exploit code has been identified at time of analysis and the issue is not listed in CISA KEV; no EPSS data or proof-of-concept reference was supplied, so the realistic threat model is a malicious or compromised privileged tenant administrator.
Insufficient isolation in Progress MarkLogic Server 12.x before 12.1.0 and 11.x before 11.3.7 allows code executed through REST Client API and XDBC evaluation paths to reach the surrounding HTTP request, response, and session state, enabling unauthorized actions or compromise of an authenticated user's session. The vendor scores this 9.6 (CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H), but the scope change and full confidentiality/integrity/availability impact only materialize when an attacker can reach an evaluation interface and a victim with an authenticated session triggers it; our assessment rates the prerequisite as authenticated (PR:L) rather than unauthenticated, so exposure is limited to deployments where these endpoints are reachable and in use. No public exploit code has been identified at time of analysis, and the flaw is not confirmed as actively exploited (no CISA KEV listing) as of this report.
Argument injection in the DDNS configuration handler of the Netcore NR268 consumer router (firmware 1.7.121109) lets an authenticated, low-privilege user manipulate the Hostname parameter sent to /router/DDNSset.cgi, which is then passed unsanitized into an execvp call and can result in arbitrary command execution on the device. Because the CVSS vector shows PR:L, the attacker must first hold valid low-level credentials or a session on the device web interface, and the router's management interface must be reachable - exposure is far greater where the admin UI is internet-facing or default credentials remain unchanged. No vendor patch has been identified at time of analysis and no public exploit code has been confirmed, though the flaw is documented in a public GitHub write-up and a VulDB entry, making details discoverable.
Authentication bypass in ((OTRS)) Community Edition 6.x and earlier lets an unauthenticated remote attacker impersonate any existing agent or customer account when HTTP Basic Authentication is the configured authentication backend. Because the CGI/mod_perl deployment maps an attacker-supplied 'Remote-User' request header into the HTTP_REMOTE_USER environment variable, OTRS treats that header value as an already-authenticated username and skips password verification, after which the Generic Interface can be accessed with the impersonated account's privileges. Exploitation is not unconditional: it requires Basic Auth to be the active backend, a CGI/mod_perl (not Mojolicious/Hypnotoad) deployment, and knowledge of a valid target username, which is why an independent assessment rates the attack complexity as High (AC:H) rather than the vendor's AC:L, and no public exploit code was identified at time of analysis. OTRS 7 and later are unaffected by this vector.
Unauthenticated remote attackers can seize global super-admin control of Rapyd Admin installations running versions 9.0.0 through 9.19.x, because the framework's setup installer seeds a hard-coded administrator account (admin@laravel, password 'admin') that grants the highest privilege level in the application. The account is silently re-created on every execution of the `rpd:make` artisan command, so an operator who previously changed the password or removed the account can have the known credential restored without warning, and a successful login bypasses per-record authorization checks and multi-tenant scoping entirely. VulnCheck disclosed the issue with full source references, no public exploit code was identified at time of analysis, and no evidence of confirmed active exploitation (CISA KEV) is present; the vendor shipped the fix in v9.20.0, which creates the first admin once with a random password and never re-seeds it.
Authentication bypass in the Grav Login OAuth plugin before 1.4.3 lets unauthenticated remote attackers derive valid credentials for OAuth-provisioned accounts and take administrative control. The plugin assigns each auto-provisioned user a deterministic '<provider>.<id>' username and an unsalted MD5 hash of the provider identifier as the password, and it leaves password login enabled; anyone who learns or guesses the provider identifier can reproduce the login. No public exploit identified at time of analysis; the vendor-assigned CVSS 4.0 score is 9.3 Critical and the assessed CVSS 3.1 vector is AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N.
Authentication bypass in ZITADEL 4.x before 4.19.2 and 3.x through 3.4.15 lets an authenticated attacker with organization-administrator permission register a SAML identity provider using their own signing key and then obtain authenticated sessions as any user linked to a different SAML provider on the same instance, including users in other organizations. The flaw stems from the SAML identity-provider intent callback verifying the assertion signature against the provider named in the request URL while resolving the user from the login intent without confirming the two match. This is a genuine high-priority vulnerability, though not a 'spray-and-pray' internet-wide threat; exploitation requires the org-admin prerequisite and that victims authenticate via SAML identity providers. No public exploit identified at time of analysis, and a vendor-released patch exists (4.19.2 and later for 4.x; a fix beyond 3.4.15 for 3.x).
Schema-level and Query/Mutation type-level @authentication directives are silently ignored for top-level @cypher fields in @neo4j/graphql versions up to and including 7.6.3, letting unauthenticated remote callers invoke those fields and run the embedded Cypher statements with the GraphQL service account's privileges, which may permit reading and modifying data. Only @authentication placed directly on the individual @cypher field is enforced, so applications relying on schema- or type-level protection are exposed while those annotating each field individually are not. No public exploit code has been identified at time of analysis, and the vendor has released 7.6.4 as the fix.
Qexo instances running versions 4.0.0 through any release before 4.2.0 expose the /api/init_step/ initialization endpoint without authentication or a setup-completion guard, allowing an unauthenticated attacker who supplies a CSRF token harvested from any public page of the instance to restart the setup flow and, in the same interaction, read back stored PROVIDER configuration such as service credentials. The CNA rates this 9.3 (Critical, CVSS 4.0) while our independent CVSS 3.1 assessment scores it AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N, reflecting that the integrity impact is a forced re-entry into initialization rather than arbitrary data modification, and that the confidentiality loss only materializes on instances that have already stored provider secrets (for example a Vercel token or image-bed credentials). No public exploit code or confirmed active exploitation was identified at time of analysis, and the issue is fixed in Qexo 4.2.0.
A container escape in sandboxed.sh through 1.3.0 lets code running inside a container workspace read and modify files on the host filesystem, defeating the isolation the product is designed to provide. The flaw stems from the tool invoking nsenter without the --root flag, so absolute paths resolved by agent-run code land on the host root rather than the container's own filesystem. Exploitation requires the attacker to already execute code inside a sandboxed.sh workspace (low privilege, local), not remote or unauthenticated access, and no public exploit code or confirmed active exploitation has been identified; the effective CVSS 3.1 vector is AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H, which materially tempers the headline 9.3 score despite the high confidentiality, integrity, and availability impact.
Remote code execution in NetEase Cloud Music for Windows 3.1.38 and 3.1.39.205426 allows an attacker with an active adversary-in-the-middle position to forge the /pc/upgrade/get upgrade-check response using a publicly known AES key over an unvalidated TLS connection, inject renderer content, and invoke the os.navigateExternal IPC bridge with a file:// URL to execute arbitrary local binaries under the logged-in user. The vendor CVSS 4.0 score is 9.2, and the independent assessment retains high confidentiality, integrity, and availability impact with no privileges or user interaction required, though AC:H reflects the network-position prerequisite; no public exploit identified at time of analysis and no CISA KEV active exploitation is indicated.
Authentication bypass in LibreNMS before 26.9.0 allows unauthenticated remote attackers to gain administrative access when the system is configured for LDAP or Active Directory authorization and the REMOTE_USER server variable is not forwarded by an upstream proxy. The flaw arises because usernames are inserted unescaped into LDAP search filters, and authentication only verifies that a matching entry exists; an attacker can supply a wildcard username via the HTTP Basic Authorization header (PHP_AUTH_USER) to match all users and groups, resulting in admin privileges. Exploitation requires the vulnerable authorization mode to be active and REMOTE_USER to be absent; installations using other auth backends or a correctly configured front-end proxy are not exploitable via this path. No public exploit code has been identified at time of analysis, and there is no confirmed active exploitation (CISA KEV).
Authentication bypass in ZITADEL's Login V2 UI (versions 3.0.0 through 4.19.1) allows unauthenticated attackers to take over user accounts, including administrators with MFA completed, by forging unsigned session cookies and guessing time-ordered session IDs around a victim's sign-in to trigger token rotation. Exploitation requires the target to run the affected Login V2 UI and the attacker to hit a narrow timing window; no public exploit code is identified at time of analysis. While the CVE carries a CVSS 4.0 score of 9.2, the independent assessment rates it with high attack complexity (CVSS 3.1 AV:N/AC:H/PR:N/UI:N), reflecting the timing prerequisite.
Server-side request forgery in Crawl4AI (unclecode) versions 0.7.0 through 0.9.3 lets unauthenticated, remote API clients force the crawler to fetch internal, private, and cloud-metadata URLs and hand back the parsed head data of those responses. The attack abuses the optional link_preview_config on /crawl requests: an attacker submits a page they control with include_external and include_patterns enabled, and the AsyncUrlSeeder component then follows links pointing at internal services or the cloud instance metadata endpoint. Exploitation requires the non-default link_preview_config feature to be active and the /crawl API to be network-reachable; the CVSS vector (PR:N, UI:N) indicates no authentication or user interaction is needed once those conditions hold. No public exploit code has been identified at time of analysis, and the issue is rated 9.2 under CVSS 4.0 (independent assessment: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:N), making this a genuinely high-priority flaw for affected deployments.
Malicious firmware updates can be installed on devices running SWUpdate through 2026.05.1 due to improper validation of X.509 certificates embedded in CMS SignedData, allowing an attacker with a signing key from a sibling intermediate CA under the same trusted root to bypass signature verification. The vulnerability is present in default builds and requires no user interaction or device privileges, but exploitation is limited to environments where the trusted root has multiple intermediate CAs. No public exploit code or active exploitation has been identified at time of analysis.
Unauthenticated SQL injection in phpMyFAQ versions prior to 4.1.9 lets remote attackers read credentials and configuration secrets, modify or delete records, or drop tables via the search parameter of the api/autocomplete endpoint. Exploitation is conditional: the backend MySQL or MariaDB instance must be running in NO_BACKSLASH_ESCAPES mode, because PdoMysql::escape() relies on backslash escaping of single quotes that that SQL mode treats literally; the payload additionally uses a Unicode right single quotation mark (U+2019) that the application normalizes to an ASCII quote after escaping has already been applied, bypassing the filter. This is a genuine high-priority issue (CVSS 4.0 base 9.2, AT:P/AC:H reflecting the non-default SQL-mode prerequisite) rather than a high-CVSS artifact, since an attacker who reaches such a deployment needs no authentication and no user interaction; no public exploit code and no confirmed active exploitation were identified at time of analysis, and the primary limiting factor is whether the target database uses the default SQL mode.
Unauthenticated account takeover in Portal del Hermano (Netherman, portaldelhermano.es) builds earlier than 2026-10-06 exploits the password-recovery endpoints POST /password/create-access and POST /password/email, which lack rate limiting and return distinct error messages acting as oracles for DNI, date of birth, phone number, and last four IBAN digits. An attacker can confirm the DNI, brute-force date of birth and IBAN digits, obtain the phone number via OSINT, and supply an attacker-controlled email to receive the password-reset link, achieving full takeover of any member account across all tenants, including profile data, full IBAN, and fiscal documents. No public exploit identified at time of analysis; disclosed CVSS 9.1 (CVSS 4.0) reflects high impact but high attack complexity (AC:H, PR:N, UI:N per independent assessment).
Account takeover of registered Thirty Bees customers is possible through an authentication bypass in PasswordController::resolveCustomer() affecting versions through 1.7.0. Unauthenticated remote attackers who know a victim's email and operate on a store with guest checkout enabled can place a guest order using that email, retrieve the resulting guest account's secure_key, and submit it as a password-reset token to reset the victim's password because the application does not distinguish guest from registered accounts. No public exploit identified at time of analysis, but the vendor has released an upstream fix commit (66410fd2a7ca9d1121a4b9551f59637ae6e4bcf1).
OTRS fails to immediately invalidate previously issued JWT authentication tokens when an agent or customer user account is deactivated or its password is changed, leaving tokens valid until their normal expiration (approximately 16 hours). An attacker who already possesses a valid token can continue accessing OTRS with the associated privileges, though customer tokens are limited to that account and agent tokens may reach broader ticket, customer, or administrative functions. The flaw provides no way to obtain a token; exploitation requires prior token possession, and the provided CVSS 9.1 (PR:N) overstates ease of attack because it does not account for this prerequisite. No public exploit identified at time of analysis.
Unauthenticated attackers can bypass security controls on REST endpoints generated by Quarkus REST Data Panache when a developer uses @MethodProperties to customize an operation without re-declaring class-level security roles, causing the generator to silently drop the security requirement and expose the endpoint. This results in potential unauthorized information disclosure or data modification, with a CVSS 3.1 score of 9.1 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N) and CWE-425. No public exploit code has been identified at the time of analysis.
Remote unauthenticated attackers can manipulate object attributes through the Apache Allura REST API in versions up to and including 1.21.0, because the API's object-binding logic (CWE-915, mass assignment) fails to whitelist which client-supplied fields may be written. The published CVSS 3.1 vector is AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N, giving a base score of 9.1 with high confidentiality and integrity impact and no availability impact, and no user interaction or credentials are required per that vector, though the practical exposure is limited by the small, largely self-hosted population of Allura instances and whether the affected write endpoints are reachable or gated behind authentication in a given deployment. Exploitation does not appear to be widespread: EPSS is only 0.29% (20th percentile), there is no CISA KEV listing, and no public exploit code was identified at time of analysis; the vendor-released fix is Apache Allura 1.22.0.
Privasys ra-tls-clients verifiers before 0.11.0 accept TEE attestation evidence without binding it to an explicitly authorized platform identity, so an unauthenticated remote attacker (CVSS vector PR:N) who can either forge attestation evidence or extract attestation secrets from a separate compromised-but-still-unrevoked Intel SGX/TDX or AMD SEV-SNP platform can present valid measurements from that other machine and impersonate the attested TLS endpoint to the verifier. Exploitation is gated by high attack complexity (AC:H, AT:P) because the attacker must actually control or forge a genuine unrevoked TEE platform, but confidentiality and integrity impact is rated High (assessed CVSS:3.1 8.1; vendor CVSS 4.0 score 9.1), and affected deployments are precisely those running the default policy with an empty platform allow-list. No public exploit code has been identified at time of analysis and the issue is not listed in CISA KEV; the flaw is fixed by ra-tls-clients 0.11.0, which introduces platform allow-listing that fails closed.
TLS certificate validation bypass in NetEase Cloud Music for Windows 3.1.38 and 3.1.39.205426 lets a network-adjacent attacker who already holds a man-in-the-middle position present a rogue CA certificate, decrypt HTTPS traffic, and capture the MUSIC_U session credential for account takeover; in-transit response modification is also possible. The client validates neither the server certificate chain nor the hostname and does not consult the Windows certificate trust store, so no victim authentication or interaction beyond normal app use is required. The reported CVSS 4.0 base score is 9.1, but exploitation is constrained by the positional prerequisite (shared Wi-Fi, rogue AP, ARP spoofing, or compromised router), and no public exploit identified at time of analysis and no CISA KEV entry has been observed.
OAuth1 request replay in Apache Allura through 1.21.0 lets a remote, unauthenticated attacker reuse a previously captured OAuth1-signed request to act with the original requester's authorization, because the platform does not adequately enforce nonce/timestamp single-use. The practical barrier is high rather than trivial: the attacker must first obtain a genuine signed request from a legitimate user - most plausibly through network interception, a man-in-the-middle position, or access to logged or leaked requests - and the replay only yields the scope of that specific captured request rather than arbitrary access, which is why the attack-complexity is assessed as high (AC:H) despite the PR:N vector. No public exploit code and no confirmed active exploitation were identified at time of analysis; EPSS is low at roughly 0.15% (4th percentile), so mass scanning is unlikely, though the vendor has published a fix in Apache Allura 1.22.0.
Open redirect flaw in the OAuth1 callback validation of Apache Allura through 1.21.0 allows unauthenticated remote attackers to redirect users to arbitrary attacker-controlled URLs during the OAuth authorization flow (CWE-601), potentially exposing authorization codes or tokens if a victim completes the flow. Exploitation is not drive-by: per the independent assessment a user must be induced to traverse the attacker-influenced OAuth request (UI:R), which lowers real-world risk below the supplied 9.1 base score; the vendor patched the issue in Allura 1.22.0, no public exploit identified at time of analysis, and EPSS is 0.15% (4th percentile), so this is assessed as a moderate real-world priority.