Password reset token exposure in phpMyFAQ before 4.1.7 enables unauthenticated account takeover by reading a publicly accessible, date-stamped tracking file. When user tracking is enabled, the application writes password reset tokens to content/core/data/trackingDDMMYYYY without access controls, allowing any network attacker to retrieve the file and replay the token against the reset API. No public exploit code or CISA KEV listing has been identified at time of analysis, but the attack path is low-complexity once the preconditions are met.
Improper authorization in Akaunting 3.1.21's BulkActions dispatcher allows authenticated low-privilege users to perform bulk operations beyond their intended access scope, potentially exposing or modifying financial records belonging to other users or organizational entities within the platform. Discovered and reported by Fluid Attacks, the flaw resides in a shared dispatcher component used across multiple modules of this open-source web-based accounting application. No public exploit has been identified at time of analysis, and a vendor-released patch is available in version 3.2.1.
Multiple cryptographic implementation defects in the libcrux Rust crates - libcrux-ecdh, libcrux-ed25519, and libcrux-psq - expose applications to availability failures and potential key-material weaknesses. libcrux-ecdh omitted length validation and clamping enforcement on X25519 secret inputs (including a broken clamping check for imported keys), libcrux-ed25519 applied a duplicated clamping step during key generation that could silently corrupt key material, and libcrux-psq panicked on AEAD decryption errors instead of returning them to callers. Vendor-released patch versions 0.0.6 (ecdh, ed25519) and 0.0.7 (psq) resolve all three defects; no public exploit code or CISA KEV listing has been identified at time of analysis.
OS command injection in the TP-Link Archer C20 v6 web management interface enables an authenticated administrator to execute arbitrary system commands by submitting crafted WAN configuration inputs that bypass insufficient input validation. Full device compromise - confidentiality, integrity, and availability - is achievable for all traffic routed through the affected device. TP-Link has released a patched firmware; no public exploit code or CISA KEV listing has been identified at time of analysis.
Arbitrary file read, write, and Python code execution in Agno's PythonTools (agno.tools.python) exposes any file accessible to the agent process via path traversal sequences injected into the file_name argument of read_file, save_to_file, or run_python_file. The root cause is use of Path.joinpath() without resolving or validating the resulting path against the intended base_dir boundary, enabling escape via sequences such as '../../../../../../etc/passwd'. No public exploit has been identified at time of analysis, but the attack is trivially constructible and can be delivered through direct tool invocation or prompt injection embedded in agent-processed content, giving any untrusted input source a path to OS-level code execution within the agent process.
Cross-tenant broken object level authorization (BOLA) in stigmem-node exposes quarantined facts across tenant boundaries on multi-tenant deployments. A tenant administrator holding only a standard tenant write capability can list, read, and admit or reject quarantined facts belonging to any other tenant on the same node via the /v1/quarantine endpoints - bypassing isolation enforced by the stigmem-plugin-multi-tenant plugin. No public exploit identified at time of analysis; no CISA KEV listing. Default single-tenant deployments are entirely unaffected regardless of version.
Two-factor authentication bypass in phpMyFAQ before 4.1.7 allows any attacker holding valid first-factor credentials to obtain a remember-me cookie prior to 2FA challenge completion, then replay that cookie to achieve full authenticated access while entirely circumventing the second-factor control. Affected deployments relying on 2FA as their primary defense-in-depth layer against credential compromise are effectively left with single-factor authentication. No public exploit code or CISA KEV listing has been identified at time of analysis, though the bypass logic is straightforward and the vulnerability is trivially reproducible given valid credentials.
SQL injection in phpMyFAQ before 4.1.7 allows authenticated low-privileged users to exfiltrate and manipulate database contents via the glossary create and update endpoints. The flaw stems from a string truncation defect that breaks SQL escaping: when an escaped string is truncated before embedding in a SQL literal, a trailing backslash can dangle, escaping the closing quote and allowing arbitrary SQL injection. No public exploit or CISA KEV listing has been identified at time of analysis, but the CVSS 4.0 score of 8.6 with high confidentiality and integrity impact reflects serious potential for sensitive data exposure.
Unauthenticated path traversal in the Simple File List WordPress plugin (versions through 6.3.11) exposes an unvalidated file-move endpoint accessible without authentication, enabling arbitrary server file reads and displacement of critical WordPress files outside the web root. Remote attackers can exfiltrate sensitive data such as wp-config.php database credentials and secret keys, and can displace core WordPress files to cause site failure or enable site takeover. No public exploit code has been identified at time of analysis, but the CISA SSVC framework rates the attack as automatable with no privileges required, elevating real-world risk despite a low current EPSS score.
Unauthenticated SQL injection in the Product Shortlist WordPress plugin (versions through 1.0.4) allows any remote attacker to query the underlying MySQL database without credentials. The CVSS:3.1 vector (AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:N) reflects trivial network exploitation with a scope change, meaning data accessible extends beyond the plugin's own tables to the shared WordPress database. No public exploit identified at time of analysis, and the EPSS score of 0.19% (9th percentile) suggests limited observed exploitation activity consistent with a niche plugin install base.
Unauthenticated SQL injection and unauthorized data destruction affect the Dinatur WordPress plugin through version 1.18. The same vulnerable request handler accepts unsanitized user input directly into SQL queries and also executes a full database table truncation with no authorization check - meaning any unauthenticated visitor can both exfiltrate WordPress database contents and irreversibly wipe the plugin's stored data. No public exploit code or CISA KEV listing has been identified at time of analysis, but the zero-barrier exploitation (no credentials, no special configuration) makes this a high-urgency finding for any site running the affected plugin.
Stored XSS in Grav Shortcode Core Plugin (prior to 6.2.2) allows an authenticated user with admin.pages permission to inject arbitrary JavaScript into page content by supplying a crafted shortcode attribute value that breaks out of an HTML attribute context. The plugin's shortcode syntax bypasses Grav's save-time XSS filter because shortcode notation never contains a literal less-than character, and multiple attribute handlers (ColorShortcode, DivShortcode, ColumnsShortcode, and others) then concatenate the raw parameter directly into rendered HTML. The stored payload executes in the browser of any visitor or administrator who subsequently views the affected page. No public exploit or CISA KEV listing is identified at time of analysis.
Wekan's outgoing webhook URL validator in versions 8.36 through 9.73 allows authenticated board members to bypass SSRF protections by supplying DNS-resolvable hostnames - such as 169-254-169-254.nip.io resolving to the cloud metadata address 169.254.169.254 - that pass the synchronous string-based blocklist check but reach blocked internal IP ranges at request delivery time. Although the pre-existing delivery-time fetchSafe guard independently blocked the reported proof-of-concept IPv4 cloud metadata target since v8.35/v8.36, the inconsistent dual-validation architecture left IPv6 address family resolution unchecked and maintained a separate, drift-prone blocklist in ssrfGuard.js that could diverge from the input-time validator across software revisions. No public exploit identified at time of analysis; vendor-released patch is confirmed as version 9.74.
SQL injection in Community by PeepSo versions up to and including 9.0.5.2 is reachable by any subscriber-level authenticated user, enabling network-based extraction of sensitive database contents with a changed scope. Classified under CWE-89, the flaw requires only low-privilege WordPress credentials and no user interaction, lowering the exploitation barrier to any registered site member. No public exploit code or CISA KEV listing is identified at time of analysis, but the minimal privilege requirement and high confidentiality impact make this a high-priority finding for sites with open registration.
Command injection in Renovate's hermit manager (versions 32.135.0 through 40.33.0) allows an attacker with repository write access to execute arbitrary commands on the machine running Renovate by supplying a maliciously named hermit dependency. The root cause is the unsanitized concatenation of user-controlled `depName` values directly into `./hermit install` and `./hermit uninstall` shell commands in `lib/modules/manager/hermit/artifacts.ts`. A detailed proof-of-concept is publicly available in GitHub Security Advisory GHSA-36j9-mx87-2cff; no confirmed active exploitation or CISA KEV listing exists at time of analysis.
Command injection in Renovate's helmv3 manager (versions 31.51.0 through 40.32.x) allows repository contributors with write access to execute arbitrary OS commands on the system running Renovate by embedding shell metacharacters in the `repository` field of a crafted `Chart.yaml` file. The unsanitized value is interpolated directly into a `helm registry login` shell invocation without applying `shlex.quote()` escaping. A detailed proof-of-concept demonstrating process termination via `kill 1` is publicly documented in GHSA-3f44-xw83-3pmg; no public exploit identified at time of analysis beyond the POC, and CISA KEV does not list this CVE.
Command injection in Renovate's npm manager (versions 35.63.0 through 40.33.0) allows attackers with repository write access to execute arbitrary commands on the CI/CD host running Renovate by supplying malicious `packageName` values via a crafted `renovate.json5` configuration file. The root cause is in `lib/modules/manager/npm/post-update/npm.ts`, where user-controlled package names and workspace keys are interpolated directly into shell commands during lock file maintenance without shell escaping via `shlex.quote()`. A working proof of concept is publicly available in the official GitHub Security Advisory (GHSA-fr4j-65pv-gjjj), and a vendor-released patch exists in version 40.33.0; no confirmed active exploitation has been reported at time of analysis.
Command injection in Renovate's gleam manager (versions 39.53.0 through 40.33.0 exclusive) allows attackers with repository write access to execute arbitrary OS commands on the machine running Renovate by committing a crafted gleam.toml containing shell metacharacters in dependency names. The depName field sourced from gleam.toml is interpolated directly into a gleam deps update shell invocation without using shlex quoting, enabling full shell escape. A detailed proof-of-concept is publicly documented in the GHSA advisory (GHSA-xjr7-3c3g-m763); no confirmed active exploitation appears in CISA KEV at time of analysis.
Command injection in Renovate's kustomize manager (versions 39.218.0 through 40.33.0) allows repository contributors with write access to execute arbitrary shell commands on the Renovate host machine by embedding crafted chart names in `kustomization.yaml` files that bypass shell escaping in `helm pull` invocations. A detailed proof-of-concept demonstrating process termination via `kill 1` against a Docker-hosted Renovate runner was published with GHSA-xv56-3wq5-9997, confirming exploitability - though no confirmed active exploitation (CISA KEV) has been observed. Exploitation is constrained by the requirement for both repository write access and the non-default `kustomizeInflateHelmCharts` postUpdateOption.
Command injection in Renovate's Gradle Wrapper update mechanism allows arbitrary code execution inside the Renovate runtime when the tool processes a repository containing a malicious `gradle-wrapper.properties`. Versions >= 32.124.0 and < 42.68.5 of renovatebot/renovate (and Mend renovate-ce/renovate-ee before 13.3.0) are affected whenever Renovate performs Gradle Wrapper artifact handling. An attacker who can introduce a crafted `distributionUrl` containing `$(...)` shell substitution into a scanned repository gains arbitrary command execution as the Renovate process, potentially exfiltrating CI/CD tokens, repository credentials, and internal network resources; a proof-of-concept screen recording demonstrating `/etc/passwd` exfiltration exists, and no public exploit identified at time of analysis for CISA KEV status.
Missing capability checks in the Advanced File Manager WordPress plugin (before 5.4.13) allow any authenticated user holding administrator-delegated file-manager access - as low as Subscriber - to read arbitrary server files including wp-config.php and to overwrite existing non-PHP files, creating a direct path to full WordPress site compromise. The CVSS 8.5 score with a scope-changed vector reflects the severity of that impact, while AC:H captures the critical prerequisite: an administrator must explicitly grant file-manager access to low-privileged roles before exploitation is possible. No public exploit has been identified at time of analysis, and a vendor patch is available in version 5.4.13.
Heap buffer overflow in FFmpeg's VC-2/Dirac RTP packetizer (libavformat/rtpenc_vc2hq.c) enables memory corruption via crafted Dirac data units across all versions prior to fix commit 1cdeb3c. The packetizer accepts an attacker-controlled data unit or fragment size and copies it into a fixed-size heap buffer without an upper-bound check, a classic CWE-122 condition with High confidentiality, integrity, and availability impact. No public exploit or active exploitation (CISA KEV) has been identified at time of analysis; an upstream fix commit is available from the FFmpeg repository.
Stack buffer overflow in FFmpeg's MPEG-PS muxer (libavformat/mpegenc.c) enables arbitrary code execution when processing crafted media input containing more streams than the muxer's fixed-size stack buffer can accommodate. Affected is all FFmpeg before commit 9d786e4; exploitation requires the victim to mux attacker-controlled input using the MPEG-PS output format, making this a file-processing attack rather than a remote network vulnerability. No public exploit code has been identified at time of analysis, but the patch commit and PR #24087 are publicly accessible, lowering the barrier for independent reproduction.
Heap buffer overflow in FFmpeg's hvcC box writer exposes any version prior to commit acf5d7c to memory corruption during HEVC muxing. When a crafted HEVC file presents more NAL units of a single type than the count field can encode, integer overflow corrupts heap memory, potentially enabling arbitrary code execution in the FFmpeg process context. No public exploit has been identified at time of analysis and no CISA KEV listing exists, but server-side transcoding pipelines ingesting untrusted HEVC content effectively elevate the local-vector risk to a network-reachable attack surface.
SQL injection in YITH WooCommerce Membership Premium versions 2.33.0 and below allows authenticated subscribers to execute arbitrary SQL queries against the underlying database. Exploitation requires only a low-privilege WordPress subscriber account - the lowest registered user role - making the attack surface broad on any membership-enabled site. The CVSS scope change (S:C) and high confidentiality impact (C:H) indicate that successful exploitation can expose data beyond the plugin itself, including sensitive user and order records stored in the WordPress database. No public exploit code or active exploitation has been identified at time of analysis.
Authorization bypass in OpenStack Aodh's alarm list API allows any authenticated user with the reader role to enumerate alarms across all projects by exploiting a logic flaw where passing `all_projects=false` removes the project-scoping key entirely rather than restricting scope - exposing trust webhook URLs, Heat signal endpoints, project IDs, and user IDs from foreign projects. The leaked webhook URLs directly enable a second-stage attack against OpenStack Watcher, whose webhook trigger endpoint has enforced no authorization since the Ussuri release (Watcher 4.0.0), permitting any authenticated user who obtains a webhook URL to start EVENT audits and execute associated action plans against arbitrary projects. No public exploit code or CISA KEV listing has been identified at time of analysis; the CVSS 4.0 score of 8.4 reflects the chained confidentiality-then-cross-system-integrity impact of these two related flaws.
Privilege escalation in FreeBSD's POSIX shared memory subsystem allows any unprivileged local user to free kernel memory that remains in active use, then access that freed memory to overwrite kernel data structures and gain root privileges. The flaw spans four active FreeBSD release branches (14.3, 14.4, 15.0, 15.1) and is triggered by two standard system calls - open(2) with O_TRUNC and fspacectl(2) - when applied to largepage-backed shared memory objects. No public exploit has been identified and the vulnerability is not listed in CISA KEV; patched release levels are available for all affected branches via FreeBSD Security Advisory FreeBSD-SA-26:44.posixshm.
Path traversal in GitPython before 3.1.58 allows remote attackers who control a Git repository to create fully initialized, attacker-controlled Git repositories at arbitrary filesystem paths outside the intended clone directory. GitPython's independent submodule implementation reads submodule names from .gitmodules without any validation - a check that core Git added years ago in response to CVE-2018-11235 - and passes the raw, traversal-containing name directly to os.makedirs() and git clone. A proof-of-concept with Docker-based reproduction is publicly documented in GHSA-hmq2-w58f-27jc; no CISA KEV listing is present at time of analysis, indicating exploitation is not yet confirmed as widespread.
Remote code execution in IBM Power Systems Firmware is possible via a stack-based buffer overflow (CWE-121) in the FSP management network protocol. An attacker holding authenticated HMC administrator credentials and adjacent-network access to the management interface can trigger the overflow to execute arbitrary code on the Flexible Service Processor, achieving full out-of-band control over all managed hardware and hosted workloads. No public exploit code and no CISA KEV listing are confirmed at time of analysis; the vendor has released a patch.
Use-after-free in the FreeBSD 15.1-RELEASE kernel's kqueue subsystem allows an unprivileged local user to corrupt kernel memory and escalate privileges to root. During a fork() system call, the kernel's knote copy routine fails to hold appropriate locks while reading knote state, and does not guard against a timer-based knote firing and being enqueued on the active list mid-copy, resulting in a double-enqueue of the new knote and subsequent list corruption. No public exploit or active exploitation has been identified at time of analysis, but the CVSS score of 8.4 (Local/No Privileges Required) reflects a realistic, high-severity local privilege escalation path on affected systems.
Privilege escalation in FreeBSD's RACK TCP stack via a use-after-free condition allows an unprivileged local user to achieve full system compromise. The RACK setsockopt(2) handler temporarily releases the connection lock while copying option data from userspace; if a concurrent context switches TCP stacks twice during this window, the stale saved pointer to the per-connection control block references freed memory while the integrity check erroneously passes. Affected versions span FreeBSD 14.3, 14.4, 15.0, and 15.1 release branches, with vendor patches confirmed via FreeBSD-SA-26:43.tcp; no public exploit has been identified at time of analysis.
IBM PowerVM Hypervisor across four firmware trains (FW950 through FW1120) exposes an integer overflow (CWE-190) in its hypervisor call interface that permits a guest partition root user to inject arbitrary quantities of data into hypervisor or co-resident partition memory. The scope change (S:C in CVSS) reflects that exploitation crosses the guest-hypervisor isolation boundary, enabling either a full platform re-IPL that terminates every co-hosted partition or silent corruption of hypervisor and partition memory. No public exploit has been identified at time of analysis, and IBM has released patches covering all affected firmware trains.
Concurrent sponsored-payment flooding in ZenHive mpp (versions 0.2.0 through 0.12.0) allows an unauthenticated remote client to drain a fee-payer wallet to zero, denying service to all legitimate sponsored-transaction users once the wallet is empty. The root cause is that MPP.Methods.Tempo.FeePayerPolicy enforces per-transaction budget ceilings but maintains no aggregate accounting across concurrent in-flight sponsorships, meaning N simultaneous sponsored requests commit N × max_total_fee against the wallet with no global cap. No public exploit has been identified at time of analysis, but the attack is mechanically trivial once the sponsor endpoint is reachable - no authentication, no cryptographic preconditions, only the ability to issue distinct signed transactions with distinct nonces within the 900-second validity window.
Privilege escalation in Splunk AI Toolkit before 6.0.0 allows any authenticated low-privileged user to operate with system-level permissions by exploiting a session key substitution flaw in the Agent Run History handler. The handler unconditionally replaces the caller's session key with a system authentication token before executing searches, completely bypassing Splunk's role-based access controls that gate elevated operations behind 'admin' or 'power' roles. This grants unauthorized read access to all indexed data, cross-user search job enumeration and deletion, and write-level system integrity impact - no public exploit has been identified at time of analysis, but the low exploitation complexity and clear mechanism description make this a high-priority patching target for Splunk enterprise deployments.
Missing authorization checks across multiple REST API handlers in Splunk AI Toolkit versions below 6.0.0 allow any low-privileged authenticated Splunk user - one who holds neither the 'admin' nor 'power' role - to start, stop, and reconfigure containers and to read or modify sensitive connection and configuration data. The vulnerability stems from CWE-862 (Missing Authorization), where the REST endpoints simply omit role-based access enforcement rather than incorrectly implementing it. No public exploit code and no CISA KEV listing have been identified at time of analysis, but the low attack complexity and the fact that it was reported by Cisco make it a credible near-term target.
Server-side request forgery in ArcadeDB before 26.8.1 allows authenticated low-privilege users to weaponize the OpenCypher LOAD CSV feature to issue arbitrary HTTP/HTTPS requests from the database server host, reaching internal network services and cloud metadata endpoints such as AWS IMDSv1. The server fetches and returns the content of these restricted destinations, enabling exfiltration of internal service data or cloud IAM credentials. No public exploit code or CISA KEV listing has been identified at time of analysis, but the CVSS 4.0 score of 8.3 with SC:H reflects meaningful lateral impact against adjacent cloud and internal infrastructure.
Cross-site request forgery in IBM Power Systems Firmware's ASMI web interface allows a network-based attacker to silently execute arbitrary administrative actions against the Flexible Service Processor (FSP) by tricking a logged-in ASMI administrator into visiting a crafted page. Affected versions span four major firmware branches - FW950, FW1060, FW1110, and FW1120 - covering a broad range of IBM Power hardware in production. No public exploit identified at time of analysis, though the high integrity and availability impact against firmware-level management infrastructure makes this a significant risk for affected organizations.
Remote unauthenticated manipulation of network traffic and DNS configuration is possible in IBM AIX 7.2 and 7.3, as well as IBM PowerVM VIOS 4.1, due to improper authentication controls (CWE-287). The CVSS 8.2 vector (AV:N/AC:L/PR:N/UI:N) confirms exploitation requires no credentials, no user interaction, and no special network positioning - making this accessible to any remote attacker with TCP connectivity to an affected system. No public exploit code has been identified at time of analysis, and CISA KEV listing has not been confirmed, but the low attack complexity and high integrity impact make this a meaningful priority for IBM Power Systems environments.
Unauthenticated remote file overwrite in IBM AIX 7.2/7.3 and PowerVM VIOS 4.1 is possible via the nimesis registration service, which fails to sanitize path traversal sequences in client-supplied input. An attacker with network access to the nimesis service can write arbitrary content to locations outside the intended directory, potentially overwriting sensitive system files such as cron jobs, init scripts, or authentication databases. No public exploit or active exploitation has been identified at time of analysis, though the CVSS vector (AV:N/AC:L/PR:N/UI:N) indicates this is straightforwardly exploitable without authentication.
Capture-replay authentication bypass in ZenHive mpp versions 0.6.1-0.6.3 allows unauthenticated network attackers to claim paid resources by reusing a legitimate customer's public TIP-20 TransferWithMemo transaction hash. When the Tempo payment method is configured with a static 'memo' value in method_config, the per-challenge attribution nonce binding that ties a specific transfer to a specific verification challenge is skipped unconditionally, and the hash credential path performs no sender or wallet-signature check - leaving only token, recipient, amount, and the static memo (itself broadcast in every unauthenticated 402 response) to match. No active exploitation is confirmed and no public exploit code has been identified; a vendor patch is available in mpp 0.6.4.
Unauthenticated remote access to NFS-exported filesystems is possible in IBM AIX 7.2 and 7.3 and IBM PowerVM VIOS 4.1 due to improper authentication (CWE-287). The CVSS vector AV:N/AC:L/PR:N/UI:N confirms that unauthenticated remote attackers on the network can exploit this with low complexity and no user interaction, gaining high-confidence read access and limited write access to exported filesystem data. No public exploit has been identified at time of analysis, and IBM has released a patch via advisory 7283858.
Case-sensitivity bypass in Grav's Apache .htaccess rules prior to version 2.0.4 allows unauthenticated remote attackers to retrieve password hashes and security configuration files by requesting uppercase or mixed-case path variants (e.g., `/USER/accounts/admin.yaml`) on case-insensitive filesystems. The protection boundary - Apache mod_rewrite RewriteRule directives blocking access to user/accounts, user/config, user/data, and user/env - fails entirely when the underlying filesystem resolves uppercase URLs to the same protected files that lowercase rules would block. No public exploit or CISA KEV listing has been identified at time of analysis, but the attack requires no authentication and no user interaction, making it trivially repeatable against eligible deployments.
Privilege escalation in Wazuh 4.9.0-4.10.3 and 4.11.0-4.14.5 allows any authenticated user holding the users_admin role to overwrite the password of protected built-in administrator accounts - including the wazuh superuser - via a logic flaw in the PUT /security/users/{user_id} REST API endpoint. The API controller passes request.get("user"), which resolves to None because that key is absent from the request object, instead of extracting the authenticated principal from the JWT token; remove_nones_to_dict() then silently strips the None before it reaches the reserved-account authorization guard in the framework layer, causing the protection check to fail open. The issue is confirmed fixed in v4.10.4 and v4.14.6; no public exploit code or CISA KEV listing has been identified at time of analysis.
Unauthenticated access to the Kafka Connect REST API in Splunk Connect for Kafka versions below 2.2.7 enables an SSRF-class attack that reconfigures the HTTP Event Collector destination to an attacker-controlled endpoint, resulting in theft of HEC authentication credentials and partial manipulation of event delivery to Splunk Enterprise. The connector's HEC endpoint validation does not enforce secure transport by default, meaning a non-TLS HTTP destination can be substituted without any authentication. No public exploit has been identified at time of analysis, but the attack requires no privileges and trivially low complexity, making any deployment with a network-exposed Kafka Connect REST API immediately actionable.
Permissive CORS misconfiguration in the Grav API Plugin (all versions prior to 1.0.0-rc.16) allows cross-origin JavaScript on any attacker-controlled website to submit a stolen JWT via Authorization or X-API-Token headers and read full authenticated API responses, enabling data exfiltration and account modification with the token owner's privileges. The wildcard Access-Control-Allow-Origin: * header was returned by default for all /api/v1 endpoints - including authenticated ones - bypassing browser Same-Origin Policy protections entirely. A secondary vector existed via URL query-string token acceptance (?token=) on general API routes, leaking JWTs into server logs and Referer headers where they could be harvested and then weaponized via the CORS bypass. No public exploit code has been identified at time of analysis, and no CISA KEV listing is present.
Arbitrary file deletion in the Atarim AI Agency for WordPress plugin (all versions through 5.1.1) allows authenticated attackers holding author-level credentials to delete any file on the underlying server via a chained directory traversal attack, with a direct path to remote code execution. The exploit chain poisons WordPress attachment metadata with a traversal path and then triggers the plugin's media replacement workflow to resolve and unlink the attacker-chosen target file. No public exploit has been identified at time of analysis, but the Wordfence advisory fully documents the two-step chain against verified source code, making independent reproduction straightforward for a moderately skilled attacker.
IBM PowerVM Hypervisor across four active firmware release trains is vulnerable to arbitrary Hostboot memory read and write via the FSP service processor mailbox interface, enabling compromise of the entire host firmware boot stack and the hypervisor subsequently loaded from it. An authenticated attacker holding FSP service-level credentials can send a specially crafted mailbox message that triggers an out-of-bounds memory operation (CWE-125), achieving full confidentiality, integrity, and availability impact against the managed POWER-series system. No public exploit code and no CISA KEV listing have been identified at time of analysis, but the scope change to the hypervisor layer makes this a critical risk for IBM POWER environments with shared or exposed FSP credentials.
Remote code execution in IBM PowerVM Hypervisor firmware versions FW950 through FW1120 allows an attacker with authenticated service-level access to the Flexible Service Processor (FSP) to execute arbitrary code in the host firmware runtime and gain full control of the managed Power system. An integer overflow (CWE-190) in the service processor mailbox interface is the root cause, with CVSS scope change (S:C) reflecting that exploitation breaks out from the FSP boundary into the broader host firmware and managed-system context. No public exploit or CISA KEV listing exists at time of analysis; however, the combination of complete C/I/A impact and scope change makes this a high-priority patch target for any organization running affected IBM Power hardware.
Full host-system memory access in IBM Power Systems Firmware allows a privileged BMC/FSP attacker to read and write arbitrary regions of host memory, yielding complete control over the host and all hosted logical partitions. Affected firmware spans multiple generations including FW1120.00, FW1110.x, FW1060.x, FW950.x, and OP940.x lines covering Power9 and Power HMC platforms. No public exploit code has been identified at time of analysis, and CISA KEV listing is absent, though the scope-changing impact (S:C) and full C/I/A triad make this a critical priority for any environment where BMC/FSP access is not tightly controlled.
Arbitrary code execution on IBM Power host systems is possible when an attacker with service processor access supplies a maliciously crafted firmware update image that bypasses boot-time image validation. Affected firmware versions span three release trains: FW1120.00, FW1110.00-FW1110.30, and FW1060.00-FW1060.80. Exploitation yields full confidentiality, integrity, and availability impact on the host, with scope change indicating the compromise extends beyond the service processor itself. No public exploit or CISA KEV listing is identified at this time.