Information disclosure in Brocade SANnav before 2.4.0b and 3.0.0 occurs because the product writes encoded passwords and authentication tokens to log files, including the SANnav supportsave diagnostic bundle. An attacker who gains access to those logs or the supportsave output can recover secrets and potentially reuse them, though the description frames this as requiring an authenticated attacker with file-level access and the assessed vector reflects a local/authenticated prerequisite; the input CVSS 4.0 vector lists PR:N, a discrepancy that should be confirmed with the vendor. No public exploit identified at time of analysis, and no CISA KEV confirmation was provided.
Security-restriction bypass in IBM DataPower Gateway 10.5.0.0 through 10.5.0.22, 10.6.0.0 through 10.6.0.10, 10.6.1 through 10.6.6, and 11.0.0.0 through 11.0.0.2 allows a remote authenticated attacker to have forged or tampered cryptographic signatures accepted by the gateway, defeating the trust checks those signatures are meant to enforce. Per the assessed vector (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N, 8.1), the attacker must already hold valid low-privilege credentials and have network reachability to the gateway, but no user interaction is required, and successful exploitation yields high confidentiality and integrity impact with no availability effect. No public exploit code was identified at time of analysis, and the issue is not confirmed as actively exploited (CISA KEV); the primary limiting factor is the authentication prerequisite, which excludes fully unauthenticated external attackers.
Deeply nested GraphQL documents can crash PHP worker processes running the webonyx/graphql-php library in versions 15.32.2 and earlier, because the recursive-descent parser has no recursion depth limit and the resulting stack exhaustion becomes an uncatchable SIGSEGV. The flaw is remotely reachable unauthenticated (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, 8.2) and the failure occurs during parsing, before query validation, depth analysis or complexity controls ever execute, so worker processes on PHP-FPM, Swoole, RoadRunner, ReactPHP or CLI runtimes are killed with no exception, log entry or recovery path. The vendor has released a fix in 15.32.3 that enables a default recursion limit of 256; there is no KEV listing and no public exploit identified at time of analysis, though the upstream advisory documents a minimal crashing payload of roughly 74 KB, which is below common HTTP body-size limits.
SQL injection in MariaDB Connector/Node.js (3.2.0 through 3.2.4, 3.3.x before 3.3.4, 3.4.x before 3.4.7, and 3.5.x before 3.5.4) allows an attacker-influenced object key containing a backtick to break out of the quoted identifier the driver builds when it expands object properties into a SQL SET clause, letting the remainder of the key be parsed as SQL so that columns the application never intended to expose can be updated or arbitrary statements appended with the database user's privileges. Exploitation is conditional and unauthenticated at the driver layer (CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H): the application must explicitly enable the non-default permitSetMultiParamEntries option and must pass object KEYS - not merely values - derived from attacker-influenced input into an expanded `SET ?` clause. No public exploit code or confirmed active exploitation was identified at time of analysis, and the limiting factors (off-by-default feature plus the need for attacker-controlled keys) are the reason the assessed vector carries AC:H; the serialized-object handling used when the option is disabled is not affected.
Out-of-bounds write in IBM DataPower Gateway lets a remote, unauthenticated attacker corrupt memory and potentially execute arbitrary code on affected gateway builds (10.5.0.0 through 10.5.0.22, 10.6.0.0 through 10.6.0.10, 10.6.1 through 10.6.6, and 11.0.0.0 through 11.0.0.2), as reflected in the CVSS v3.1 vector AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H and a base score of 8.1. The PR:N/UI:N components mean no credentials or user interaction are required, so any network-reachable instance on an affected build is exposed, though the AC:H rating indicates exploitation depends on satisfying a non-trivial condition - most likely a specific input state, parsing path, or memory layout - to escalate the write beyond a crash into reliable code execution. Vendor patch data indicates a fix is available from IBM, and no public exploit code or confirmed active exploitation was identified at time of analysis; the EPSS score was not provided in the underlying intelligence, so no probabilistic exploitation estimate can be quoted.
Authorization bypass in Apache DolphinScheduler before 3.4.3 allows an authenticated user to alter task definitions, together with their upstream dependencies, in projects they are not authorized to access. The attacker pairs a projectCode they legitimately belong to with a task-definition code belonging to a different project when calling the /dolphinscheduler/projects/{projectCode}/task-definition/{code}/with-upstream endpoint, because the handler never confirms that the task code actually belongs to the supplied project. Impact is confined to integrity and availability of workflow definitions (assessed vector I:H/A:L, C:N), so unauthorized projects may have their pipelines silently rewritten or disrupted; no public exploit code has been identified at time of analysis and EPSS is only 0.18% (7th percentile), making this a genuine but moderate-priority issue rather than an urgent emergency.
Authenticated users of Apache DolphinScheduler can stop or create savepoints for task instances belonging to projects they are not authorized to access, because the project-scoped task-instance API fails to enforce per-project authorization. The issue affects all releases before 3.4.3 and is reachable over the network via the /dolphinscheduler/projects/{projectCode}/task-instances/{taskInstanceId}/stop and /savepoint endpoints. The independent assessment rates this CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:L (the vendor-published score is 8.1), meaning a valid authenticated account plus knowledge or enumeration of a target task instance ID is required - anonymous exploitation is not possible - and the practical impact is an integrity and limited availability concern in multi-tenant or multi-project deployments where project isolation is a security boundary; no public exploit code has been identified at time of analysis, EPSS is a low 0.16% (5th percentile), and the CVE is not in CISA KEV. This is a genuine but moderate-priority issue rather than a top-tier emergency.
Man-in-the-middle interception of TLS-protected communications in IBM Guardium Data Protection 12.2.2 is possible because the product does not properly validate certificates (CWE-295), allowing an attacker positioned on the network path to read and modify traffic that the software treats as trusted. The flaw is unauthenticated (CVSS 3.1 AV:N/AC:H/PR:N/UI:N) but not trivially automatable, since the attacker must first achieve an active interception position - the reason CISA SSVC rates exploitation as none and automatable as no, and the reason the assessed impact is limited to confidentiality and integrity (C:H/I:H/A:N) rather than availability. EPSS is very low at 0.14% (3rd percentile), no public exploit code has been identified, and the issue does not appear in CISA KEV; IBM has released a vendor patch, so the practical risk for most deployments is remediation-tracking rather than emergency response.
Malcolm, CISA's open-source network traffic analysis platform, ships an nginx-based reverse proxy whose Lua RBAC authorization layer and nginx's request routing normalize crafted URL paths differently, letting an authenticated user at any privilege level slip past role-based restrictions and reach administrative or role-gated endpoints. Rated CVSS 8.1 (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N), the flaw yields high confidentiality and integrity impact against every RBAC-protected path - file upload, the PHP server, htadmin, and authentication management interfaces - while requiring only network reachability plus a valid low-privilege account (PR:L), with no user interaction. There is no public exploit identified at time of analysis and no confirmed CISA KEV listing, but the low credential bar and the sensitivity of the exposed interfaces make this a genuine priority rather than a CVSS-inflated paper risk.
Stack-based buffer overflow in IBM DataPower Gateway firmware tracks 10.5.0.0 through 10.5.0.22, 10.6.0.0 through 10.6.0.10, 10.6.1 through 10.6.6, and 11.0.0.0 through 11.0.0.2 allows an authenticated attacker holding administrative-level privileges to execute arbitrary code on the gateway. Exploitation is network reachable but gated by high attack complexity (AC:H), meaning the attacker depends on conditions they cannot fully control, such as a specific memory layout or parsing state, so a single deterministic trigger is not guaranteed; the assessed vector is CVSS:3.1/AV:N/AC:H/PR:H/UI:N/S:U/C:H/I:H/A:H. There is no public exploit identified at time of analysis and the issue is not reported as confirmed actively exploited (CISA KEV); IBM has released patched builds via its support advisory.
Cross-site scripting in Progressive Robot hMailServer 6.3.2 through 6.3.5 can let a remote, unauthenticated attacker execute script in the webmail origin with a victim's session if the victim decrypts a crafted S/MIME (from 6.3.2) or OpenPGP (from 6.3.4) message and opens a text/html attachment in a new tab. The webmail is only served when the non-default REST API is enabled, and the attack requires user interaction, so realistic risk is materially lower than the 8.0 CVSS score. No public exploit was identified at time of analysis, and the vendor released v6.3.6 as a fix.
Checksum verification bypass in the Go toolchain (cmd/go) versions 1.27.0 before 1.27.2 and before 1.26.9 allows a malicious Go project with a forged golang.org/toolchain go.sum entry, combined with a user-selected attacker-controlled GOMODPROXY, to substitute a tampered toolchain that cmd/go would otherwise reject. Exploitation requires local access and user interaction (CVSS 7.8, AV:L/UI:R) and two attacker-controlled preconditions the victim must opt into: operating inside the crafted project and using the malicious proxy. No public exploit code identified at time of analysis, and EPSS is 0.13% (2nd percentile), indicating low predicted exploitation likelihood.
Go toolchain (cmd/go) releases before 1.26.9 and 1.27.0 through 1.27.1 can be induced to substitute an attacker-controlled module in place of the bundled golang.org/fips140 dependency when a developer builds a project that ships a bogus golang.org/fips140 definition and connects the build to an attacker-operated GOMODPROXY. Successful abuse yields high confidentiality, integrity, and availability impact within the build environment because the toolchain fails to properly validate the module's integrity check value (CWE-354), so untrusted module content is accepted in place of the trusted bundled copy. Exploitation is conditional rather than widespread: it requires both an attacker-crafted project and explicit developer action to use a non-default, attacker-chosen proxy (CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H, base 7.8), and no public exploit code or confirmed active exploitation has been identified at time of analysis, with EPSS at roughly 0.11% (1st percentile). Vendor patches are available in Go 1.26.9 and 1.27.2.
Local privilege escalation in IBM Guardium Data Protection 12.2 allows an attacker who already holds low-privilege local access on the appliance to recover the root password through the pkcrypto passkey component and thereby obtain full root control, because that component combines weak cryptographic protection with a hard-coded recovery key (CWE-798). The exposure is confined to a local attack path (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H), meaning no network or unauthenticated remote vector exists and the attacker must first establish a foothold on the appliance - realistically an insider or someone who has already compromised the host by other means. There is no public exploit identified at time of analysis, CISA SSVC rates exploitation as none and the flaw as non-automatable while assigning total technical impact, and EPSS is 0.11% (1st percentile), so this is a serious but not urgent technical issue.
Local privilege escalation to root in the IBM Guardium Data Protection 12.2 patch installer, where a signature-verification bypass lets an attacker substitute a tampered or unsigned patch artifact that the installer then executes with root privileges. Exploitation requires local access to the appliance with at least low-level privileges (PR:L) and the ability to stage a malicious patch and win a timing window during installation; the assessed vector is AV:L/AC:H with a resulting base score of roughly 7.0, slightly below the vendor-published 7.8. No public exploit code and no confirmed active exploitation were identified at time of analysis, and the vendor has published a patch; this is a credible but not emergency-tier priority given the local-only attack path, prerequisite foothold, and race-condition reliability constraints.
Local privilege escalation in Guardium Data Protection 12.2 allows an authenticated low-privileged local user to abuse the cp_wrapper component to gain root privileges and access or modify sensitive system files. The issue is not remotely exploitable (AV:L) and requires a pre-existing local foothold (PR:L), so it is most relevant to insider threats, compromised low-privileged accounts, and post-compromise escalation; no public exploit identified at time of analysis, and no CISA KEV confirmation was provided. IBM has released a vendor patch, and this is a credible but not emergency-tier priority.
Path traversal in decompress-zip through 0.3.3 allows a crafted ZIP archive to write files outside the intended extraction directory when a victim or application uses the package to extract it, potentially enabling arbitrary file overwrite, application compromise, or code execution in environments with a useful writable sibling path. The assessed vector is local (AV:L) with required user/application interaction (UI:R), no network exposure, and no special authentication (PR:N), so exploitation requires the attacker to deliver the archive into an extraction workflow; RCE is conditional, not guaranteed. No public exploit identified at time of analysis, and no CISA KEV active-exploitation report was provided; the overall risk is real but conditional rather than an urgent mass-exploitation threat.
Local privilege escalation in GlavSoft TightVNC Server for Windows before 2.8.88 allows an authenticated local user to execute arbitrary code with SYSTEM privileges by planting a malicious screenhooks32.dll or screenhooks64.dll where the TightVNC service will load it. DynamicLibrary::init() and ThemeLib call LoadLibrary() with bare file names and no LOAD_LIBRARY_SEARCH_* flags, so the service follows the default DLL search order and picks up an attacker-controlled copy from a writable directory that sits ahead of the legitimate file, such as an installation directory left with weak ACLs. Exploitation is constrained: it requires a pre-existing local session on the affected host plus a writable directory earlier in the search order, so it cannot be triggered remotely or unauthenticated; no public exploit identified at time of analysis.
Local privilege escalation in TightVNC Server for Windows before 2.8.88 allows an authenticated low-privileged user to tamper with the service's IPC channel, potentially gaining SYSTEM-level access, disclosing session data, or causing denial of service. The flaw stems from a named shared memory segment in the Global\ namespace created with a NULL DACL and a predictable time-seeded name, enabling any local process to open and modify pipe HANDLE values used between the TightVNC service and its desktop server process. No public exploit code has been identified at time of analysis, and exploitation requires a local authenticated session on a host actively running the vulnerable service as SYSTEM.
Incorrect default permissions in the Windows installer for Progressive Robot hMailServer 6.0.0 through 6.3.5 let any local authenticated user read the mail server's stored messages, logs, temporary files, and built-in database, because the data, log, temp, database and event folders and the hMailServer.INI file inherit the local Users group's read access from the default installation path under Program Files. The readable INI exposes the database password, which is protected only by the machine's DPAPI key and can therefore be unsealed by any local account - handing full control of an external database - while the built-in database's account password hashes become readable whenever the hMailServer service is stopped; the Linux AppImage 6.3.0 through 6.3.5 has a parallel weakness in per-user data folder permissions. Exploitation requires an untrusted local account able to sign in to the host, no public exploit was identified at time of analysis, and the vendor has released 6.3.6 as the fix.
Heap-based buffer overflow in hMailServer 6.0.0 through 6.3.3 on Windows lets a local interactive user - with no hMailServer account - crash the mail server service and potentially execute code in its LocalSystem security context by passing an over-long hexadecimal string to the unauthenticated COM method Utilities.BlowfishDecrypt, which copies input of arbitrary length into a fixed 255-byte heap buffer. Because exploitation requires an existing local login (CVSS AV:L/PR:L) and the only confirmed outcome is a service crash, this is an access-gated local privilege-escalation/denial-of-service issue rather than a remotely wormable emergency; no public exploit identified at time of analysis. Vendor fix is available in hMailServer 6.3.4, which contains the reference commit 135a1908ff820ed587d5f180f98c53bd010d8931.
Local privilege escalation in hMailServer 6.0.0 through 6.3.5 on Windows allows an authenticated local user-one with an interactive logon but no hMailServer credentials-to abuse COM objects to read and write arbitrary files as the service account and send mail as any sender. The service registers its COM classes without DCOM launch or access permissions and calls CoInitializeSecurity with a null security descriptor, so any user logged on at the console or over Remote Desktop can activate hMailServer.Message, Attachments, and hMailServer.FetchAccount objects. If the service runs as LocalSystem (the default), arbitrary file write can lead to code execution as SYSTEM; no public exploit identified at time of analysis.
Local privilege escalation in Progressive Robot hMailServer 6.3.4 and 6.3.5 on Linux allows an attacker who already has code execution as the unprivileged hmailserver service account to run arbitrary code as root via the live-update apply helper (hmailserver-update). The vulnerability requires the live-update path unit to be active, which is the default for the project's .deb and .rpm packages and applies to AppImage installations run under that unit, but does not affect non-Linux builds. No public exploit code has been identified at time of analysis, and the vendor has released version 6.3.6 to address the issue.
Uncontrolled resource consumption in the Strawberry GraphQL library's legacy graphql-ws subscription handler (versions 0.312.3 up to but not including 0.327.2) allows an unauthenticated remote client on a persistent WebSocket connection to fill that connection's configured subscription slots with one-shot subscriptions that have already completed, so later legitimate operations on the same connection are rejected with 'Subscription limit reached'. The impact is limited to the attacker's own connection and to availability (CVSS 3.1 base 3.7, AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:L), requiring the non-default max_subscriptions_per_connection cap to be explicitly configured and the legacy graphql-ws protocol to be in use; the modern graphql-transport-ws protocol and deployments without the cap are unaffected. No public exploit code has been identified at time of analysis and the issue is not known to be actively exploited, but a vendor fix is available in 0.327.2.
Denial of service in IBM DataPower Gateway 11.0.0.0 through 11.0.0.2 allows an attacker who already holds a valid low-privileged account to trigger improper memory allocation during key derivation, corrupting memory and rendering the gateway unavailable (CVSS 7.7, AV:N/AC:L/PR:L/UI:N/S:C/C:N/I:N/A:H). Because the vector requires PR:L, the flaw is not exploitable pre-authentication, and its scope-changed rating reflects that the availability impact can extend past the initially compromised security boundary; there is no confidentiality or integrity impact. Patch available per vendor advisory and no public exploit identified at time of analysis, so risk is concentrated in unpatched appliances where an attacker can obtain even limited authenticated access.
Path traversal in Brocade Active Support Connectivity Gateway (ASCG) before 3.5.0 lets an unauthenticated attacker write arbitrary files to restricted host locations by embedding traversal sequences in entries of a support-bundle archive that ASCG ingests from remote endpoints, with the potential for follow-on remote code execution if a written path is later executed or loaded. The attack is not internet-facing: exploitation is scored AV:A (adjacent network) with the CVSS 4.0 AT:P precondition, meaning the attacker must already control or be able to man-in-the-middle support bundles originating from a compromised or spoofed ingesting endpoint. No public exploit code has been identified at time of analysis and the CVE is not listed in CISA KEV, so the practical risk is a moderate, conditional one that depends on an attacker holding a management-network position and a compromised ingestion source rather than an urgent, broadly exploitable issue.
Red Hat OpenShift Container Platform 4 ships by default with an insights-operator-gather ClusterRole that grants the operator's 'gather' service account cluster-wide get/list access to core-group secrets, with no namespace or resourceNames scoping. An authenticated actor who can create or influence a pod specification in any namespace - and set 'serviceAccountName: gather' on it - can mount that service account and read every secret in every namespace, making this a privilege-escalation and cluster-wide credential-disclosure primitive (CWE-269, CVSS 7.7, scope-changed confidentiality-only). Exploitation requires a pre-existing foothold with pod-create or service-account-binding rights (PR:L), so it is not remotely exploitable by an unauthenticated user; no public exploit code has been identified at time of analysis, and the over-broad grant exists out of the box rather than behind a non-default feature flag.
Authentication bypass in Brocade Fabric OS versions before 9.2.2d running on the MXG610 extension switch lets a network-adjacent attacker reach an unprotected endpoint in the web management interface's Single Sign-On workflow and obtain administrative access to the device. No credentials or user interaction are required (PR:N/UI:N), although High attack complexity (AC:H) means conditions outside the attacker's control must line up for reliable exploitation, and exposure is confined to MXG610 deployments with a reachable management interface. No public exploit code has been identified at the time of analysis, and there is no confirmation of active exploitation.
Local privilege escalation in IBM Guardium Data Protection 12.2 lets an already-authenticated, high-privileged user gain root-level command execution by feeding attacker-controlled arguments to the SUID-root ssh_config_wrapper component. Because the component runs with effective UID 0, successful argument injection converts a privileged-but-limited foothold into full root control of the appliance, with high impact to confidentiality, integrity, and availability. This is a moderate-severity escalation issue rather than a remote-takeover emergency: the attacker must already hold a high-privileged account and be able to invoke the vulnerable binary locally, and no public exploit code has been identified at time of analysis. A vendor patch is available, though the specific fixed build is not enumerated in the supplied data.
Cross-site request forgery in Pydantic AI's development web chat UI allows any website a developer happens to visit while `Agent.to_web()` or `clai web` is running to fire forged chat requests at the loopback-hosted server, causing the served agent to start a run and execute its tools with the privileges and credentials of the local process. The pre-patch `/api/chat` endpoint performed no Content-Type validation, so a malicious page can deliver a JSON body cross-origin using a CORS-safelisted content type (text/plain, multipart/form-data, application/x-www-form-urlencoded, or none at all) without triggering a preflight; binding the server to localhost - the default - provides no protection because a browser page can reach 127.0.0.1. Realized damage is bounded by the tools the agent exposes, and `requires_approval=True` does not help because the endpoint trusts approval decisions relayed by the client; exploitation is unauthenticated (CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:H/A:L, 7.6) but requires the developer to browse an attacker-controlled page while the dev UI is running, and there is no public exploit code identified at time of analysis. Vendor-released patches are available in 1.107.4 and 2.28.0.
Unauthenticated attackers with adjacent-network reachability can push an unauthorized Certificate Authority certificate into the TLS client trust store of Brocade ASCG (all versions before 3.5.0), because the product's management interface accepts CA certificates without authentication. A successful injection lets an on-path adversary Man-in-the-Middle the outbound TLS connections ASCG establishes to the managed Fibre Channel switches and peer nodes it administers, producing high confidentiality and integrity impact (CVSS 7.6; CVSS:4.0/AV:A/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:N; independent CVSS:3.1/AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:N). Because the vector is PR:N the certificate submission is unauthenticated, but exploitation still requires the attacker to be logically adjacent to the management plane (AV:A) and, for the MITM payoff, to hold an on-path position between ASCG and its outbound peers (AT:P) - no public exploit code and no CISA KEV entry were identified at time of analysis, so this is a genuine but moderate-priority issue rather than an urgent emergency.
Stored cross-site scripting can occur in applications generated by JHipster before generator-jhipster 9.4.0 and react-jhipster 1.1.0 when an authenticated user stores attacker-controlled Blob content and a companion ContentType in a Blob-bearing entity and a privileged user later opens it through the generated UI. The assessed CVSS is 7.6 with vector CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:C/C:H/I:L/A:N, reflecting authenticated write access, required victim interaction, and conditional exploitability that depends on the generated application's CSP and target-browser Blob handling. No public exploit identified at time of analysis; vendor-released patches are generator-jhipster 9.4.0 and react-jhipster 1.1.0.
Two-factor authentication bypass in pH7Builder (pH7 Social Dating CMS) before 18.5.0 lets attackers who already know a valid account password brute-force the 6-digit TOTP code submitted to VerificationCodeFormProcess.php without any attempt throttling, ultimately taking over member, affiliate, or administrator accounts. Exploitation is second-order: it requires prior credential compromise (PR:L) plus 2FA enabled on the target account, and success depends on racing the OTP rotation window (AT:P), but when those conditions hold nothing stops unlimited guesses. No public exploit code or confirmed active exploitation has been identified at time of analysis; the vendor has shipped a fix in version 18.5.0 that reuses the existing per-IP login-attempt lockout.
Blind SQL injection in the Appsbd Vitepos / vitepos-lite WordPress plugin, affecting versions through 3.6.1, allows an attacker who already holds a high-privilege account on the WordPress site to inject SQL through an improperly neutralized plugin input and extract database contents one inference at a time. The primary barrier is the authentication prerequisite (CVSS PR:H) - exploitation is remote over the network with low complexity and no user interaction, but anonymous internet attackers are excluded, and the attacker must first compromise or legitimately hold an administrative-level session. Patchstack reported the issue with a CVSS 3.1 score of 7.6; the independent assessment narrows the realized impact to loss of confidentiality (AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:N/A:N), and no public exploit code or confirmed active exploitation was identified at time of analysis.
Concurrency slot exhaustion in Pydantic AI 2.10.0 through 2.52.x lets streamed model requests permanently occupy slots in a shared ConcurrencyLimiter, progressively starving later requests until a service that reuses one long-lived limiter can no longer make model calls - an availability-only denial of service (C:N/I:N/A:H) with no confidentiality or integrity impact. The issue is remotely reachable and unauthenticated (PR:N), but our independent assessment rates it AC:H rather than GitHub's AC:L because exploitation requires a specific deployment pattern: model requests wrapped in ConcurrencyLimitedModel (or covered by limit_model_concurrency) that share a single long-lived limiter across streamed responses; agent-level max_concurrency and all non-streaming model requests are explicitly unaffected, and a short-lived or per-request limiter never accumulates enough leaked slots to deadlock. Slots are leaked by early stream termination, cancellation, consumer exceptions, or even full stream_text() consumption with debounce_by=0.1, and the vendor-released fix in 2.53.0 is available; no public exploit code identified at time of analysis.
Stack-exhaustion denial of service in msgpack5, the MessagePack v5 implementation for Node.js and browsers, affects all versions prior to 6.1.0. When an application uses msgpack5's streaming decoder and feeds it a single chunk containing many small, individually valid MessagePack values, the decoder recursively re-invokes itself once per remaining value, so recursion depth grows with the value count and eventually overflows the JavaScript call stack - crashing or stalling the process or stream. The flaw is remotely reachable without authentication or user interaction (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, 7.5), though it is confined to services that expose the streaming decoder to untrusted input; the fix is the vendor-released version 6.1.0, and no public exploit code or confirmed active exploitation was identified at time of analysis.
Decoding a truncated map32 MessagePack value in msgpack5 (npm) before 6.1.0 causes the decoder to read the four-byte length field of a five-byte header that has not fully arrived, producing an out-of-bounds buffer read that throws RangeError instead of IncompleteBufferError; the unauthenticated network-vector flaw (CVSS 3.1 7.5, AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H) can abort a request, stream, or worker in applications that use IncompleteBufferError to decide whether to wait for more bytes. Impact is denial of service only - the underlying buffer implementation bounds-checks the read, so there is no adjacent-memory disclosure and no path to code execution. No public exploit code or confirmed active exploitation was identified at time of analysis, and the vendor has released a fixed version (6.1.0).
A heap buffer overflow in IBM DataPower Gateway (affected firmware levels 10.5.0.0 through 10.5.0.22, 10.6.0.0 through 10.6.0.10, 10.6.1 through 10.6.6, and 11.0.0.0 through 11.0.0.2) allows remote attackers to crash the service and cause a denial of service; the CVSS vector (AV:N/AC:L/PR:N/UI:N) indicates no authentication, privileges, or user interaction are required, so any reachable affected gateway is a candidate target. Impact is confined to availability - the vendor reports no confidentiality or integrity loss and no code execution. No public exploit code has been identified at time of analysis and there is no confirmed active exploitation (not in CISA KEV); this is a genuine availability risk but not a top-tier emergency, with practical exposure limited by how broadly the gateway's data-plane and management interfaces are reachable.
Malformed GeoJSON Polygon or MultiPolygon values cause MariaDB Connector/Node.js - every release before 3.2.5, 3.3.4, 3.4.7 and 3.5.4 - to write uninitialized Node.js heap memory into database values. When a calling application passes a non-array 'ring' (an object exposing a numeric length, a string, or a number) through the binary-protocol execute() or batch() APIs, the encoder sizes a Buffer.allocUnsafe() allocation from that ring's length but the writing loop skips the reserved bytes, so the entire buffer is sent and persisted in the database, potentially carrying other users' content, session material, database credentials or TLS key material that can then spread to backups and replicas. The published CVSS 3.1 vector (7.5, AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N) treats this as a trivially reachable unauthenticated disclosure, but independent review rates attack complexity High because exploitation additionally requires the attacker to influence the geometry value the application passes to the connector, the application to actually use binary-protocol prepared statements or batches with geometry rather than the unaffected text-protocol query() path, and the ability to read the persisted value back; no public exploit code has been identified at time of analysis and there is no CISA KEV entry.
Denial of service in IBM DataPower Gateway allows a remote, unauthenticated attacker to crash or degrade an affected gateway by sending crafted input that the product fails to validate correctly, resulting in an out-of-bounds write in memory. All four supported release trains are affected: 10.5.0.0 through 10.5.0.22, 10.6.0.0 through 10.6.0.10, 10.6.1 through 10.6.6, and 11.0.0.0 through 11.0.0.2. The CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, score 7.5) confirms the attack is launched over the network with no authentication and no user interaction, and that the sole impact is availability - there is no confidentiality or integrity loss, so the risk is bounded to service outage on gateways whose vulnerable interface is reachable from untrusted networks. No public exploit code has been identified at time of analysis, no EPSS score was supplied, and CISA KEV does not list this CVE.
Denial of service in IBM DataPower Gateway allows a remote, unauthenticated attacker to crash or exhaust the message-processing service by sending crafted messages whose length field is mishandled during memory reallocation. All firmware levels in the ranges 10.5.0.0 through 10.5.0.22, 10.6.0.0 through 10.6.0.10, 10.6.1 through 10.6.6, and 11.0.0.0 through 11.0.0.2 are affected; the CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, score 7.5) indicates no authentication, user interaction, or non-default configuration is needed, and the only impact is loss of availability. There is no confirmed active exploitation (not in CISA KEV) and no public exploit code identified at time of analysis, but because exploitation only requires network reachability to the gateway's message interface, internet- or partner-facing appliances should be treated as the highest-priority remediation targets.
Unauthenticated remote attackers can exhaust system resources on IBM DataPower Gateway firmware releases 11.0.0.0 through 11.0.0.2, degrading or halting availability of the gateway's traffic and management services. The flaw is rated CVSS 7.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H), meaning no credentials, user interaction, or special positioning are required once the affected interface is reachable, but impact is confined to availability - there is no data disclosure, tampering, or code execution component. No public exploit code has been identified and there is no confirmed active exploitation (not listed in CISA KEV); the practical risk is therefore a bounded denial-of-service exposure that is highest where DataPower interfaces are internet-facing or unsegmented, and much lower where they sit behind network controls and rate limiting.
Spoofing of trusted event sources in Microsoft Azure Event Grid is possible because the service fails to enforce an authorization check on event publishing and routing operations (CWE-862). Exploitation requires the attacker to already hold a valid, low-privilege identity in the Azure environment (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:N/I:H/A:N), so this is not a remote-unauthenticated attack: an authorized principal who can reach the Event Grid endpoint over the network can push events across an authorization boundary they should not cross, undermining the integrity of event-driven workflows. Impact is integrity-only with no confidentiality or availability loss, scope is assessed as changed, no public exploit code has been identified at time of analysis, and because Event Grid is a Microsoft-managed cloud service the fix is applied server-side by the vendor rather than by customers themselves.
Authentication bypass in IBM Security Verify Access (10.0 through 10.0.9.2) and IBM Verify Identity Access (11.0 through 11.0.3), including their container editions, allows a remote, unauthenticated attacker to defeat security restrictions and gain unauthorized access to functionality that should be protected. With a CVSS 3.1 base score of 7.5 (AV:N/AC:L/PR:N/UI:N, C:N/I:H/A:N), the flaw is network-reachable, low complexity, and requires no credentials or user interaction; the impact is rated as high on integrity, meaning an attacker can alter protected data or configuration through the bypass. The only practical precondition is that the affected authentication or management interface be reachable by the attacker. No public exploit code has been identified at time of analysis and there is no confirmed active exploitation (not listed in CISA KEV); a vendor patch is available per IBM advisory node 7291628.
Authentication bypass in IBM Security Verify Access 10.0 through 10.0.9.2 and IBM Verify Identity Access 11.0 through 11.0.3 (including the container editions) allows a remote, unauthenticated attacker to defeat the product's authentication control entirely, with no user interaction required. The 7.5 CVSS score reflects a high integrity impact only - the vector (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:N) assigns no confidentiality or availability impact - so the practical effect is that an attacker can perform authenticated actions with the privileges of a legitimate session while exposing no data directly. IBM has published a vendor fix via support node 7291628, and no public exploit code has been identified at time of analysis; this is a credible operational priority but not a confirmed emergency.
Path traversal (CWE-22) in IBM Security Verify Access 10.0 through 10.0.9.2 and IBM Verify Identity Access 11.0 through 11.0.3 allows a remote, unauthenticated attacker to write arbitrary files on the affected appliance or container by supplying crafted path components. Per the CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:N, 7.5) exploitation is network-reachable, low-complexity and unauthenticated, with high integrity impact and no confirmed confidentiality or availability impact - a file-write primitive rather than a file-read or denial-of-service bug. No public exploit identified at time of analysis and the CVE is not listed in CISA KEV; the vendor has released a patch and the sole reference is IBM's advisory at https://www.ibm.com/support/pages/node/7291628. Exposure is bounded by network reachability: instances sitting behind reverse proxies or confined to internal/DMZ segments that attackers cannot reach are not directly exploitable.
Unauthenticated path traversal in IBM Guardium Data Protection 12.2 allows a remote attacker who can reach the GIM (Guardium Installation Manager) file-upload interface to escape the intended upload directory and write arbitrary files onto a Collector host. The assessed vector (AV:N/AC:L/PR:N/UI:N, integrity impact) confirms that no credentials or user interaction are required against that endpoint, so any 12.2 Collector whose GIM upload service is reachable is exposed until patched; the practical attack surface is driven almost entirely by network segmentation of GIM. No public exploit code has been identified at time of analysis, and exploitation has not been confirmed as active - impact is file-system write access on the Collector, scoring CVSS 7.5.
Denial of service in IBM DataPower Gateway affects releases 10.5.0.0 through 10.5.0.22, 10.6.0.0 through 10.6.0.10, 10.6.1 through 10.6.6, and 11.0.0.0 through 11.0.0.2, where crafted network input reaching the vulnerable parsing path overflows a buffer and can crash the affected service. Per the CVSS vector (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H), exploitation requires no authentication and no user interaction, so any internet- or partner-facing listener that processes malformed input is exposed; impact is availability-only - a crash/restart condition rather than data exposure or code execution. No public exploit code has been identified at time of analysis and the flaw is not listed as confirmed actively exploited (CISA KEV), so this is a genuine but bounded operational risk that depends primarily on whether affected listeners are reachable from untrusted networks.
Denial of service in IBM Guardium Data Protection 12.2.2 allows an unauthenticated remote attacker with network reachability to the edge-controller gRPC service to terminate the edge-controller process by supplying malformed task data that trips an unchecked type assertion. The assessed vector (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, score 7.5) confirms no authentication, no user interaction, and no special timing are required once the port is exposed, and impact is confined to availability - the process crashes unexpectedly with no data disclosure or modification. There is no public exploit identified at time of analysis and no confirmed active exploitation (the CVE is not in CISA KEV), so the practical exposure is governed by reachability: deployments that keep the edge-controller gRPC endpoint on isolated management networks are not exposed to arbitrary remote attackers, while internet- or broad-LAN-accessible instances are at meaningful disruption risk. The primary limiting factor is that the gRPC port must be reachable, making this a genuine but bounded-impact availability issue.
IBM DataPower Gateway can be driven into a denial-of-service condition by a remote, unauthenticated attacker who supplies input that reaches a flawed sorting comparator, causing the comparator to violate strict-weak-ordering and triggering an out-of-bounds memory access (CWE-787) that can crash or destabilize the affected service. All builds in the 10.5.0.0-10.5.0.22, 10.6.0.0-10.6.0.10, 10.6.1-10.6.6, and 11.0.0.0-11.0.0.2 ranges are affected; the impact is confined to availability (CVSS 3.1 base 7.5, AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H) with no confidentiality or integrity effect and no privilege gain. At time of analysis there is no confirmed active exploitation (not in CISA KEV) and no public exploit code identified, and IBM has published a fix via its support advisory, so this should be treated as a patching-priority availability risk rather than an emergency.