Remote information disclosure in IBM AIX 7.2/7.3 and PowerVM VIOS 4.1 stems from an integer underflow in the kernel's IPv4 IP-options parser (CWE-125 out-of-bounds read), letting a remote attacker leak sensitive memory contents by sending crafted IP packets. IBM (the reporting vendor) rates it CVSS 9.4, and a patch is available. There is no public exploit identified at time of analysis and it is not listed in CISA KEV.
Unauthenticated remote code execution in InternLM LMDeploy (versions 0.9.2 through 0.15.x) allows a remote attacker to execute arbitrary code inside an inference engine process when disaggregated (P2P) serving is enabled. The disaggregation peer-connector deserializes attacker-influenced ZMQ messages with Python pickle before any type validation, and the /distserve/p2p_initialize and /distserve/p2p_connect endpoints require no authentication in the default configuration (api_keys=None). Reported by VulnCheck with a vendor advisory and a public patch; no public exploit or CISA KEV listing identified at time of analysis, though the code path and fix are publicly documented on GitHub.
Improper access control in Splunk Enterprise (versions below 10.4.2, 10.2.6, 10.0.9, and 9.4.14) allows a party holding only an embedded report token to download the dispatch archive of an embedded report's search job, exposing session material that grants access to sensitive data and the ability to affect platform integrity. The embedded-report authorization flow fails to block dispatch-archive download requests before the archive begins streaming, so the check is effectively bypassed (CWE-284). No public exploit has been identified at time of analysis, but the 9.4 CVSS score and unauthenticated network vector make this a high-priority patch.
Improper access control in Splunk Enterprise (versions below 10.4.2, 10.2.6, 10.0.9, and 9.4.14) lets anyone holding an embedded report token download the underlying search job's dispatch archive over the REST API, recover session material, and impersonate the report owner. Because embedded reports are designed to be shared with unauthenticated viewers, an attacker can read all data the owner can access and, if the owner holds the admin role, perform administrative actions. No public exploit is identified at time of analysis, but the CVSS 9.4 rating and unauthenticated network vector make this high priority.
Authenticated remote code execution affects Talebook (MyBooks/PoxenStudio) ebook management server versions 3.41.2 and earlier, where an administrator can inject arbitrary Python into the auto-generated auto.py settings module via a crafted SOCIAL_AUTH key name. Because the server imports auto.py as a live module and can auto-restart to reload it, injected statements execute with the service account's privileges, enabling data disclosure, file modification, persistence, or service disruption. No public exploit is identified at time of analysis; the fix is version 3.42.0, and chained authorization/registration weaknesses noted by the vendor could lower the effective privilege barrier.
Session material exposure in Splunk Enterprise (versions below 10.4.1, 10.2.6, 10.0.9, and 9.4.14) allows an unauthenticated attacker who can view the HTML source of any page that embeds a Splunk report to harvest session credentials and reach the underlying search data with high confidentiality and integrity impact. The flaw stems from the dispatch archive download path failing to enforce the embedded-report authorization boundary, so archived search-job data carries sensitive session material. Rated CVSS 9.4 and reported by Cisco; no public exploit is identified at time of analysis and it is not listed in CISA KEV.
Server-Side Request Forgery in the Grav API plugin (grav-plugin-api) for Grav CMS before 1.0.6 lets a caller holding the api.webhooks.write permission register webhook URLs that the server dereferences with no protocol allow-listing. Because WebhookController.php validates the URL only with FILTER_VALIDATE_URL and WebhookDispatcher.php builds cURL without CURLOPT_PROTOCOLS/CURLOPT_REDIR_PROTOCOLS, an attacker can force the server to fetch file://, gopher://, dict://, and private/link-local targets, exfiltrating local files and internal HTTP responses including cloud metadata credentials. There is no public exploit identified at time of analysis and it is not in CISA KEV, but the root cause and fix are confirmed in the vendor advisory and commit.
Broken access control across all four first-party TinaCMS production media adapters (next-tinacms-s3, next-tinacms-dos, next-tinacms-azure, next-tinacms-cloudinary) allows any authenticated CMS editor to write or delete storage objects outside the operator-configured mediaRoot boundary. The handlers pass attacker-controlled query parameters (req.query.key for presigned S3 PUTs, req.query.media for DELETEs) directly to storage SDK operations without validating that the resulting key falls within the permitted path prefix, giving editors write and delete authority over the entire storage bucket or container reachable by the deployment's IAM or service credential. Vendor-released patches exist in all four packages; no public exploit or CISA KEV listing has been identified at time of analysis.
Heap buffer overflow in FFmpeg's RIST protocol reader (libavformat/librist.c) lets a remote RIST sender corrupt heap memory in any FFmpeg instance ingesting an async:rist:// stream, enabling potential remote code execution or crash. The librist_read() function ignores the caller's buffer size and copies the entire received RIST payload into a destination buffer that the async wrapper sizes smaller than the packet, so an oversized payload writes past the buffer end. Reported by VulnCheck and fixed upstream in commit 1c10bcc; no public exploit is identified at time of analysis and it is not listed in CISA KEV.
Authentication bypass in the Cudy WR3000 2.0 router (firmware before 2.5.24) lets unauthenticated remote attackers forge valid JWT tokens by reusing a hard-coded HMAC signing secret embedded in the firmware image, granting illegitimate access to the device's Mosquitto MQTT broker and mesh-networking control interface. Because the secret is static across all units of this model, any attacker who obtains one firmware image can authenticate against every deployed device. There is no public exploit identified at time of analysis, but the flaw is trivially exploitable once the secret is extracted, and the network-reachable, no-privilege CVSS 4.0 score of 9.3 (VC:H/VI:H) reflects full read/write access to mesh control.
Unauthenticated administrative account takeover in vsDesk (a Russian-market IT service-desk/help-desk platform) lets a remote attacker abuse insecure deserialization of application configuration data to point the login flow at an attacker-controlled LDAP server and provision a new admin account. Rated CVSS 4.0 9.3 (critical) with an unauthenticated network vector, it was researched and disclosed by Kaspersky (klsecservices); a detailed public write-up exists (KLSA-00296), though no CISA KEV listing or confirmed weaponized exploit is present in the data, so this is best characterized as no public exploit code identified at time of analysis despite a thorough advisory.
OS command injection in the Cudy WR3000 2.0 router allows an authenticated network attacker to execute arbitrary commands with root privileges by exploiting the mesh MQTT command interface. The sync_command binary passes unsanitized attacker-controlled input directly to a shell execution sink in command.lua, and the vulnerable command execution path is enabled by default. No public exploit code or CISA KEV listing has been identified at time of analysis, but the low attack complexity and root-level impact make this a high-priority patch target for any deployment using the mesh networking feature.
Authentication and authorization bypass in the cluster-proxy-addon component of Red Hat's Multicluster Engine for Kubernetes (MCE) lets an unauthenticated attacker who can reach the user-facing proxy route manipulate URL path segments to proxy requests to arbitrary services on any managed cluster. Because the flaw is a server-side request forgery (CWE-918) with a changed CVSS scope, a request to the hub's exposed route pivots into internal services on downstream clusters that were never meant to be externally reachable, enabling information disclosure and potential further compromise. No public exploit identified at time of analysis, and the issue is not listed in CISA KEV; the CVSS base score is 9.3 (critical).
SQL injection in the J-BusinessDirectory extension for Joomla (by cmsjunkie.com) before version 6.2.3 lets remote attackers inject arbitrary SQL through the trips search feature, where user-supplied search keywords and the ORDER BY sort clause are concatenated directly into a query. The CVSS 4.0 vector (AV:N/AC:L/PR:N/UI:N, score 9.3) indicates unauthenticated, low-complexity network exploitation with high confidentiality, integrity, and availability impact. There is no public exploit identified at time of analysis and it is not listed in CISA KEV, but SQL injection in a public search endpoint is trivially weaponizable.
SQL injection in the YOOtheme Zoo content-construction extension for Joomla allows unauthenticated remote attackers to inject arbitrary SQL through the filter_type and type_filter request parameters, which are concatenated directly into a.type comparison clauses in ItemController::element() without quoting or escaping. All Zoo versions before 4.1.64 are affected, and because no authentication or user interaction is required, an attacker can read or manipulate the Joomla database over the network. No public exploit has been identified at time of analysis, but the trivial exploitation profile (CVSS 9.3) makes weaponization straightforward.
TNC policy server impersonation in IBM AIX 7.2/7.3 and IBM PowerVM VIOS 4.1 lets a network-positioned attacker forge the Trusted Network Connect policy server and tamper with the traffic between it and the endpoint, because the client fails to properly validate the server's certificate (CWE-295). An attacker who can intercept the session can present a fraudulent certificate, read limited data and alter integrity-critical policy exchanges. IBM has released a fix; no public exploit identified at time of analysis, and the issue is not listed in CISA KEV.
SQL injection in the NGG Smart Image Search WordPress plugin (all versions before 4.0.0) lets remote attackers inject arbitrary SQL through the plugin's search functionality without authentication. Because the CVSS vector marks privileges as none and user interaction as none, any visitor can reach the vulnerable query, and the scope-changed high-confidentiality rating (9.3) reflects potential exposure of the entire WordPress database, including credentials and secrets. No public exploit was identified at time of analysis, and the flaw is not listed in CISA KEV; it was reported by Patchstack.
Stored cross-site-scripting-to-RCE in SiYuan (the open-source personal knowledge-management/note app) before v3.7.4 lets an attacker-crafted document execute arbitrary OS commands on a victim's machine. Document metadata fields (bookmark, alias, memo, and an alternate name field) are concatenated unescaped into the aria-label attribute of the file-tree 'move/link to' path-picker tooltip; a double quote breaks out of the attribute and injects an inline event handler. Because every SiYuan Electron window runs with nodeIntegration:true, contextIsolation:false and no CSP, the handler reaches require('child_process') and achieves command execution the moment the victim hovers the malicious entry. There is no public exploit identified at time of analysis and it is not in CISA KEV.
Stored cross-site scripting escalating to arbitrary OS command execution affects the SiYuan personal knowledge-management application through version 3.7.3. Because the '((' block-reference autocomplete popup renders a candidate block's name, alias, and memo fields into HTML without escaping, an attacker who controls those metadata fields can plant a self-firing payload that runs the moment a victim types '((' and surfaces the crafted block; SiYuan's Electron shell runs with nodeIntegration enabled, contextIsolation disabled, and no CSP, so the injected script reaches require('child_process') and executes host commands. There is no public exploit identified at time of analysis and it is not listed in CISA KEV, but the XSS-to-RCE chain is fully described by the VulnCheck advisory and vendor GHSA.
Unauthenticated SQL injection in the Total Donations WordPress plugin (versions <= 2.0.5) lets remote attackers inject arbitrary SQL via unsanitized input without any authentication, exposing the WordPress database to disclosure. The CVSS 9.3 rating with a changed scope reflects that a successful attacker can read sensitive data (user credentials, PII, donation records) beyond the plugin's own boundary. No public exploit has been identified at time of analysis, and EPSS/KEV data were not provided.
Unauthenticated SQL injection in the Nikstore Core WordPress plugin (versions <= 1.5) lets remote attackers inject arbitrary SQL through unsanitized input reaching a database query, per a Patchstack report classified under CWE-89. Any exposed WordPress site running the plugin is reachable without credentials, and the changed CVSS scope reflects the database being a distinct security authority from the web application. No public exploit identified at time of analysis, and the issue is not listed in CISA KEV.
Unauthenticated SQL injection in the Maps Marker Pro WordPress plugin (all versions up to and including 4.32) lets remote attackers inject arbitrary SQL against the site database without any credentials or user interaction. Reported by Patchstack and rated CVSS 9.3, the flaw exposes the backing WordPress database to unauthenticated extraction. No public exploit has been identified at the time of analysis and the CVE is not listed in CISA KEV.
Anonymous read/write/federation access to Stigmem nodes (versions before 0.9.0a2) is possible when operators disable authentication and bind the node to a non-loopback address, effectively bypassing authorization for the anonymous identity. Reported by VulnCheck and disclosed via GHSA-fp6w-8wpg-74g5, it affects the eidetic-labs stigmem-node package (pip). No public exploit identified at time of analysis, and no EPSS or CISA KEV signal is available; exploitation is gated on a deliberate operator misconfiguration rather than a default-config weakness.
OS command injection in Dell OpenManage Enterprise prior to version 4.7.0 enables low-privileged remote attackers to execute arbitrary operating system commands on the underlying server. The CVSS vector (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H) confirms the flaw is network-exploitable with low complexity, requiring only a low-privilege account - a minimal barrier in enterprise environments where many users hold valid credentials. No public exploit code or CISA KEV listing has been identified at time of analysis; Dell has released a patch in version 4.7.0 per advisory DSA-2026-359.
Command injection in IBM AIX 7.2 and 7.3 and IBM PowerVM VIOS 4.1 enables an unauthenticated attacker with adjacent network access to execute arbitrary OS commands, achieving full compromise of confidentiality, integrity, and availability. The adjacent-network attack vector (AV:A) limits exposure to attackers already positioned on the same local network or VLAN as the affected system, which is a meaningful constraint in typical enterprise deployments of these platforms. No public exploit code and no active exploitation via CISA KEV have been identified at time of analysis; IBM has released a patch via its support portal.
Path-traversal-driven remote code execution in Wazuh 4.3.0 through 4.14.5 (and 5.0.0-beta1/beta2) lets a cluster peer that holds the shared Fernet key overwrite security-sensitive files on the master node. Because the non-merged branch of process_files_from_worker() in master.py trusts the peer-supplied file_path from files_metadata.json without confining it under the directory implied by cluster_item_key, an attacker can replace /var/ossec/etc/ossec.conf to configure root-executed commands and gain code execution on the next service reload. No public exploit identified at time of analysis, but a vendor fix and the fixing commit are public, making the root cause easy to study.
Sensitive-key exposure in IBM AIX 7.2/7.3 and PowerVM VIOS 4.1 Network Installation Management (NIM) leaks intermediate certificate authority (CA) private keys inside a publicly downloadable update file, letting remote attackers forge trusted certificates and bypass security restrictions. Rated CVSS 9.1, the flaw impacts confidentiality and integrity (C:H/I:H) but not availability, and IBM has released a fix. There is no public exploit identified at time of analysis and it is not listed in CISA KEV, but the ease of obtaining a live CA key makes this a high-priority credential-hygiene issue.
Operating-system command execution in the Splunk MCP Server app (all versions below 1.2.1) lets a user already holding the Splunk 'admin' role run arbitrary commands on the underlying host by abusing unsafe deserialization in the credential-management component. Because the app trusts stored data and deserializes it without verifying the object type, a privileged Splunk user can escalate from application-level administration to full host compromise. No public exploit identified at time of analysis, and it is not listed in CISA KEV; a vendor patch is available.
Privilege escalation in the search-v2-operator (Red Hat Advanced Cluster Management / OpenShift search component) allows a user with edit rights on the Search Custom Resource to abuse unvalidated CR fields (imageOverride, arguments, environment variables) to load an attacker-controlled container image or mount arbitrary secrets. Because the search ServiceAccount holds broad impersonation permissions, controlling that pod pivots into full cluster compromise. No public exploit is identified at time of analysis, and it is not listed in CISA KEV; the CVSS 9.1 rating reflects the scope change from a namespaced CR editor to the entire cluster.
Mutual-TLS authentication bypass in the Netty network framework (io.netty.handler.ssl.SslClientHelloHandler) lets an unauthenticated remote attacker reach routes protected only by per-SNI clientAuth=REQUIRE. A ClientHello whose 4-byte handshake header is split across TLS records trips a wrong-offset bounds check, throwing IndexOutOfBoundsException and calling select(ctx, null), which falls back to the default SslContext (clientAuth NONE/OPTIONAL) instead of the SNI-matched context. Fixed in 4.1.137.Final and 4.2.17.Final; no public exploit identified at time of analysis, but the CVSS 4.0 base score is 9.1 (Critical).
Remote code execution in IBM AIX 7.2/7.3 and PowerVM VIOS 4.1 is achievable by an unauthenticated attacker on the same network segment via command injection triggered by crafted IPv6 Router Advertisement messages. The vulnerability (CWE-269, Improper Privilege Management) implies the RA processing subsystem executes injected commands with elevated OS-level privileges, enabling full confidentiality, integrity, and availability compromise of the affected host. No public exploit has been identified at time of analysis; IBM has released a patch per advisory https://www.ibm.com/support/pages/node/7283858.
Authentication bypass via WebAuthn assertion replay in phpMyFAQ affects all versions up to and including 4.1.5, fixed in 4.1.7. Because prepareForLogin generates a login challenge but neither WebAuthn controller writes the mutated key object back to the database, the anti-replay check is short-circuited by its own null guard, so an attacker who captures one successful WebAuthn assertion can resubmit it indefinitely to log in as that user with no hardware key or user interaction. There is no public exploit identified at time of analysis and it is not listed in CISA KEV, but the root cause is well documented in the vendor advisory.
Two-factor authentication bypass in phpMyFAQ before 4.1.7 lets an attacker who already knows a valid password defeat the TOTP second factor through unbounded brute force. Because the five-attempt failure counter is stored per session and reset on every successful password re-authentication (CWE-307), an attacker can loop - grab a fresh session cookie, re-enter the known password, and get five more guesses - until the 6-digit TOTP is guessed, yielding full account takeover. Reported by VulnCheck with no public exploit identified at time of analysis and not listed in CISA KEV.
Federation peer-registration authentication bypass in stigmem-node 0.9.0a1 lets a network-positioned attacker enroll a rogue federation peer without the required administrator out-of-band fingerprint approval, granting read access to and tampering ability over federation traffic. Reported by VulnCheck and tracked under CWE-295, it is rated CVSS 4.0 9.1 but carries meaningful exploitation prerequisites (high attack complexity plus a present attack requirement). No public exploit identified at time of analysis, and it is not listed in CISA KEV; fixed in 0.9.0a2.
Man-in-the-middle exposure in LINE's Armeria (armeria-xds module) versions prior to 1.41.0, where TLS peer (certificate) verification on xDS-managed upstream connections can be silently disabled, causing the client to trust unauthenticated or spoofed upstreams. Any attacker positioned on the network path between an armeria-xds client and its xDS-configured upstream service can intercept, read, and tamper with traffic that operators believe is protected by TLS. Reported by LY-Corporation; no public exploit identified at time of analysis and the issue is not listed in CISA KEV.
Cleartext exposure of federation traffic in eidetic-labs stigmem-node (versions before 0.9.0a2) occurs when operators explicitly disable mTLS while binding federation endpoints to non-loopback addresses, allowing an on-path attacker to intercept and tamper with inter-node federation data. Reported by VulnCheck and tracked as an information-disclosure issue (CWE-319), it carries a CVSS 4.0 base score of 9.1 driven by high confidentiality and integrity impact. No public exploit identified at time of analysis, and it is not listed in CISA KEV.
Improper authorization in the Ceph Monitor subscription handler lets any authenticated CephX user holding only `mon allow r` read the entire config-key (kv) store by sending a single crafted MMonSubscribe message, affecting Ceph releases prior to 20.2.4 (Squid) and 19.2.6 (Reef). Because that store holds OSD LUKS disk-encryption passphrases and, on cephadm-managed clusters, the SSH private key cephadm uses to reach every host as root, a low-privileged read-only account can escalate to full cluster and host compromise. There is no public exploit identified at time of analysis, though the vendor fix includes a QA reproduction test exercising the kv: subscription path.
Unauthenticated PHP object/global variable injection in the Broken Link Checker WordPress plugin before 2.4.12 allows attackers to overwrite arbitrary PHP global variables via unfiltered query variables on sites using plain permalinks, escalating to remote code execution when a classic (non-block) theme is active. The flaw is reachable by any unauthenticated visitor (CVSS 3.1 base 9.0, PR:N) but requires the AC:H preconditions of plain permalinks plus a classic theme, so it is not universally exploitable. No public exploit has been identified at time of analysis, and the EPSS probability is low (0.22%, 13th percentile), indicating no observed mass exploitation yet.
Stack-based buffer overflow in IBM AIX 7.2 and 7.3, and IBM PowerVM VIOS 4.1 allows remote authenticated attackers to execute arbitrary code on affected systems. The flaw (CWE-121) is reachable over the network with low attack complexity and requires only low-privilege credentials, yielding complete confidentiality, integrity, and availability impact. No public exploit code or CISA KEV listing has been identified at time of analysis, but the 8.8 CVSS score and confirmed RCE potential make patching urgent for organizations running IBM POWER infrastructure.
Stack-based buffer overflow in IBM AIX 7.2 and 7.3 and IBM PowerVM VIOS 4.1 enables remote authenticated attackers to execute arbitrary code, achieving full compromise of confidentiality, integrity, and availability. The CVSS 8.8 score reflects low-complexity network exploitation requiring only low-privilege authentication (PR:L), making it accessible to any credentialed user on the system. No public exploit identified at time of analysis, but a vendor-released patch is available from IBM.
Remote code execution in Splunk Enterprise allows any authenticated low-privilege user - one who does not hold the 'admin' or 'power' role - to execute arbitrary code by submitting crafted XML content to the Splunk Web Manager Configuration endpoint. The vulnerability combines two weaknesses: server-side evaluation of manager XML expressions without sufficient input restrictions (CWE-94), and a missing capability check on the configuration route that should gate such access. No public exploit has been identified at time of analysis, but the low attack complexity and low privilege requirement make this a high-priority remediation target across Splunk Enterprise 9.4.x, 10.0.x, 10.2.x, and 10.4.x branches.
Weak bcrypt work factor in Rockwell Automation OTTO Fleet Manager exposes stored password hashes to accelerated offline brute-force attacks when an attacker gains access to an unencrypted system backup. All versions covered by CPE cpe:2.3:a:rockwell_automation:otto®_fleet_manager:* are affected. An adjacent-network attacker with low-level access who obtains a system backup could recover credentials more easily than a properly configured bcrypt implementation would permit, potentially enabling privilege escalation or lateral movement within the OT environment. No public exploit code or CISA KEV listing has been identified at time of analysis.
Arbitrary command execution against IBM AIX 7.2, 7.3 and PowerVM VIOS 4.1 is possible by an unauthenticated attacker on the same network segment who can deliver DHCP option values containing unescaped shell metacharacters to the system's DHCP client. The DHCP client fails to neutralize these metacharacters before passing option data to shell processing (CWE-78), enabling full confidentiality, integrity, and availability compromise of the target host. A vendor patch is available via IBM support, and no public exploit or CISA KEV listing has been identified at time of analysis.
OS command injection in IBM AIX 7.2 and 7.3 and IBM PowerVM VIOS 4.1 enables an unauthenticated attacker with adjacent network access to execute arbitrary operating system commands without any user interaction. The flaw (CWE-78) resides in a network-exposed component that fails to neutralize shell metacharacters before incorporating user-supplied input into OS commands, granting full confidentiality, integrity, and availability impact. No public exploit has been identified at time of analysis, but a vendor patch is available via IBM support advisory.
OS command injection in IBM AIX 7.2/7.3 and IBM PowerVM VIOS 4.1 allows an adjacent network attacker without credentials to execute arbitrary operating system commands. The flaw (CWE-78) arises from improper neutralization of special elements passed to OS command constructors, exposing the full confidentiality, integrity, and availability of affected hosts. No public exploit or CISA KEV listing has been identified at time of analysis, though a vendor patch is available via IBM Support.
Remote code execution in Splunk Enterprise allows any authenticated low-privilege user (without admin or power roles) to upload a malicious knowledge bundle via the REST API and force distributed search to execute it across search peers, achieving full C/I/A compromise. The root cause is a missing authorization check - the REST API endpoint for knowledge bundle uploads does not enforce the required edit_dist_peer capability, and distributed search blindly honors caller-supplied bundle selections. No public exploit or CISA KEV listing has been identified at time of analysis, but the low authentication bar (any valid Splunk account suffices) and network-accessible attack surface make this a high-priority patching target in enterprise environments.
Remote command injection in Termix, a web-based SSH server management platform, allows any authenticated user with file-manager access to execute arbitrary commands on a managed SSH host prior to version 2.3.2. The archive creation endpoint in src/backend/ssh/file-manager.ts passes user-supplied file basenames directly to GNU tar without an end-of-options delimiter, enabling an attacker to supply filenames matching tar checkpoint options (e.g., --checkpoint=1, --checkpoint-action=exec=<cmd>) that tar interprets as directives rather than operands. No public exploit code has been identified at time of analysis and the CVE is not listed in CISA KEV, but the exploitation technique is well-documented and the CVSS score of 8.8 reflects genuine high-priority risk for all pre-2.3.2 deployments.
Unauthenticated remote code execution in ICEcoder 8.1 allows any network-reachable attacker to run arbitrary OS commands as the web-server user by exploiting a three-weakness chain in the terminal endpoint: an authentication bypass triggered by supplying a password parameter, a CSRF skip triggered by any non-empty csrf value, and unsanitized command passthrough directly into PHP's proc_open(). The attack requires a single crafted HTTP POST request and no prior credentials. A public proof-of-concept appears available via a GitHub Gist reference in the VulnCheck advisory. No active exploitation is confirmed by CISA KEV at the time of this analysis.
Arbitrary code execution in Splunk AI Toolkit below 6.0.0 is achievable by any user holding the Splunk 'power' role, through loading a crafted model file that embeds malicious pickle payloads inside sparse matrix data. The toolkit's model codec deserializes this data without sanitizing or restricting the embedded pickle content, granting the attacker code execution as the Splunk server process. No public exploit or CISA KEV listing has been identified at time of analysis, but the RCE impact with a low-privilege network vector makes this a high-priority patch target for any Splunk deployment running the AI or Machine Learning Toolkit.
Remote code execution in Splunk Enterprise allows any authenticated low-privileged user - one without the 'admin' or 'power' role - to inject and execute arbitrary code on the Splunk platform instance via the Splunk Web Manager Configuration REST API endpoint. The root cause is a missing permission check on the REST API path governing manager configuration writes, meaning Splunk evaluates attacker-supplied configuration values without verifying the caller holds the required role. Exploitation yields full confidentiality, integrity, and availability impact on the Splunk instance; no public exploit has been identified at time of analysis.