In Capsule 0.13.0 through 0.13.6, the Tenant validating admission webhook's hostnameRegexHandler.OnUpdate passes the new and old Tenant objects in the wrong order, so an update is validated against the previously stored AllowedHostnames.Regex rather than the newly submitted one; a cluster administrator can therefore persist a malformed regex that the webhook should have rejected. Because the Ingress hostname validator in validate_hostnames.go discards the regular-expression compilation error and treats every hostname as unmatched, the affected tenant's Ingress create and update operations are subsequently blocked until an administrator repairs the Tenant spec. The issue requires Tenant update privileges (PR:H in the assessed CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:C/C:N/I:N/A:H vector), is confined to tenants that use IngressOptions.AllowedHostnames.Regex, and no public exploit code has been identified at time of analysis; it is fixed in 0.13.7.
Capsule 0.13.0 through 0.13.6 accepts a malformed ForbiddenAnnotations.Regex on Tenant updates because the update webhook validates ForbiddenLabels.Regex for both the labels and annotations checks, letting a privileged tenant administrator persist a bad regular expression that later panics Capsule's forbidden-metadata validator and blocks namespace creation and updates for the affected tenant. Exploitation requires authenticated, high-privileged access - a Tenant update permission (tenant owner or cluster admin with Tenant write access, CVSS PR:H) - and the impact is a tenant-scoped denial of service with no confidentiality or integrity effect (AV:N/AC:L/PR:H/UI:N/S:C/C:N/I:N/A:H, CWE-20). There is no public exploit identified at time of analysis, no CISA KEV listing, and the issue is fixed in version 0.13.7, making this a genuine but low-urgency availability bug rather than a high-impact priority.
Denial of service in AnyIO versions prior to 4.14.2: the process-pool implementation behind to_process.run_sync() starts worker processes with stderr wired to a pipe that the parent never drains, so a worker that writes more than the roughly 64 KiB pipe buffer to sys.stderr blocks on write() before it can return its stdout protocol response, leaving the awaiting process-pool call hung indefinitely. Applications are exposed only if they actually use AnyIO's process pool and run worker functions whose stderr volume can be driven by untrusted or faulty input - for example verbose tracebacks or logging produced by attacker-influenced data - while code paths that never touch to_process, or that already close/redirect worker stderr, are unaffected. Impact is limited to availability (hang), with no confidentiality, integrity, or code-execution effect, and per the vendor and independent assessment this is a local, low-privilege-scoped flaw (CVSS:4.0 AV:L/PR:L and CVSS:3.1 AV:L/PR:L, A:H); no public exploit code exists and no active exploitation has been confirmed (no public exploit identified at time of analysis).
Stored cross-site scripting in the All Bootstrap Blocks WordPress plugin (all versions through 1.3.31) lets an authenticated user with Contributor-level access or above plant a malicious block attribute that the plugin writes into HTML tag-name position without escaping, so arbitrary JavaScript runs in the browser of anyone who later views the affected content. Publicly available exploit code exists, and the vendor's own advisory is linked from WPScan, but there is no evidence of active exploitation (no CISA KEV entry) and the plugin has a small deployment footprint, making real-world impact narrower than the headline CVSS 6.8 (Medium) suggests. Because exploitation requires a low-privilege account plus a second user triggering the view, this is best treated as a Contributor-gated stored XSS rather than a remotely exploitable, unauthenticated flaw.
Stored cross-site scripting and unauthorized content tampering in the King Addons for Elementor WordPress plugin (all versions before 51.1.81) allow authenticated users with contributor-level access or above to overwrite the Elementor content of arbitrary posts and pages, including those owned by administrators, because the template-import feature performs no object-level authorization check. A widget setting is written to the page and rendered without escaping, so the injected JavaScript executes in the session of any user who later views the tampered page. Publicly available exploit code exists and the vendor has released a fixed version (51.1.81); there is no indication of confirmed active exploitation (CISA KEV) at time of analysis.
Local file inclusion that can escalate to PHP code execution affects the All Bootstrap Blocks WordPress plugin in versions up to and including 1.3.31, allowing an authenticated user with contributor-level access or higher to direct the plugin to include arbitrary local files at render time and read their contents. Code execution occurs only where an includable local file already contains PHP, and exploitation additionally requires the plugin's non-default Lightspeed subsystem to be explicitly enabled, so a stock installation is not exposed. No CISA KEV entry has been recorded for this issue, but publicly available exploit code exists and a proof-of-concept is referenced by WPScan, making the risk material for sites that have enabled Lightspeed and grant contributor accounts.
Concurrent downgrade requests against Paymenter 1.5.6 and earlier can double-spend account credit: any authenticated customer with an active, downgradable service can fire parallel requests that each pass the unpinned Service::upgradable check and each credit the same balance, yielding N refunds for one downgrade. The inflated balance is real, spendable store credit, so an attacker can immediately settle invoices or provision services at the operator's expense; no public exploit code or confirmed active exploitation was identified at time of analysis, and the independent assessment rates this a moderate-severity business-logic flaw rather than a system-compromise vector, since it requires a qualifying service, the default credits-on-downgrade setting, and a race window that must be won. Vendor-released patch: 1.5.7.
Local information disclosure in microsandbox before 0.5.10 exposes sandbox secret values - host-side API keys, tokens and environment secrets - to any other local user or co-resident process on the same host, which can simply read them from the sandbox process command line while the sandbox is running. The Rust SDK serializes NetworkConfig secret values into the --network-config argument and passes per-sandbox secrets through repeated --env arguments, keeping them visible in /proc/<pid>/cmdline on Linux and process listings on macOS for the entire sandbox lifetime, with no code execution inside the sandbox and no access to the spawning user's session required. Rated CVSS 3.1 6.5 (AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:N/A:N, medium), the flaw is only exploitable on shared/multi-tenant hosts, CI runners and developer systems where an adversary holds a separate local account or co-tenant process, and only leaks data when secrets were actually configured; no public exploit code has been identified at time of analysis, and the issue is fixed in 0.5.10.
Web cache poisoning in the Breeze Cache WordPress plugin before 2.5.15 allows unauthenticated remote attackers to append tracking-related query parameters that the plugin still caches responses for but omits from the page-cache key, causing a page rendered in the attacker's own request context to be stored under the clean URL's cache entry and served to every subsequent visitor. Exploitation requires only that page caching be active in its default configuration - no authentication, no user interaction, and no special setup (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N, 6.5) - and publicly available exploit code exists; the issue is not confirmed as actively exploited (not in CISA KEV). Impact is bounded to low confidentiality and integrity effects, exposing one visitor's rendered context to others and altering cached page content until the entry expires or is purged.
Missing ownership verification in the RestroPress WordPress plugin through 3.4.6 lets remote, unauthenticated attackers enumerate which orders are in a recoverable state and inject arbitrary notes into other customers' orders simply by supplying an order identifier to the payment-recovery flow. The flaw is an IDOR-class authorization bypass (CWE-639) with a CVSS 3.1 score of 6.5 on the vector AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N, so no credentials, privileges, or user interaction are required and default configurations are exploitable wherever the plugin's payment/order functionality is in use. Publicly available exploit code exists and the issue was reported by WPScan; impact remains confined to order-state disclosure and note injection (no code execution, no full order-data exfiltration, no destruction), so real-world risk is moderate rather than critical.
Cross-site scripting in the MdPreview component of md-editor-v3 (versions up to and including 6.5.3) lets remote attackers execute JavaScript in the host application's origin by supplying Markdown that contains a crafted fenced-code language string, and because the host can persist that Markdown it escalates to stored XSS for every user who later views the content. User interaction is required - a victim must render the malicious Markdown - and the shipped XSSPlugin does not prevent exploitation, since it only filters existing html_block and html_inline tokens and cannot inspect the renderer-generated HTML that the highlight callback produces. Scored CVSS 3.1 6.1 (AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:N); publicly available exploit code exists (a working payload is included in the upstream GHSA advisory), no confirmed active exploitation (no CISA KEV listing), and the vendor-released patch fixes the issue in version 6.5.4.
AnyCable (anycable-go) through 1.6.14 ships its telemetry subsystem enabled by default with a hardcoded public auth token ('secret' in telemetry/config.go) and derives its cluster fingerprint by hashing the full configuration file plus raw os.Args, which can include values passed via --secret, --jwt_secret, and --http_rpc_secret; an adversary who can MITM or DNS-hijack telemetry.anycable.io can intercept the outbound telemetry POST, and because the client's token is publicly known it offers no protection. The transmitted clusterFingerprint is a SHA256 digest rather than raw plaintext, so meaningful secret recovery additionally depends on low-entropy secrets, and the required network-interception position makes this a moderate-to-low real-world risk best characterized as a hardcoded-credential and telemetry-exposure design flaw rather than a directly exploitable credential leak. Exploitation status: no public exploit code identified at time of analysis and no confirmed active exploitation (CISA KEV); the GHSA advisory documents only a conceptual DNS-spoofing reproduction. Vendor-released patch: 1.6.15.
Signed-request replay in AnyCable's Pusher-compatible REST API (version 1.6.14 and earlier) lets anyone who obtains a single legitimate signed POST request resend it with a completely different body, because the handler folds the caller-supplied body_md5 into the HMAC string but never computes or compares the digest of the body actually received. The same code path omits any auth_timestamp freshness check, so a captured auth_signature remains valid indefinitely. Successful abuse causes the server to broadcast attacker-chosen event content to arbitrary channels within that signed request's application context, enabling forged server-side events, application-state corruption, or phishing messages delivered to WebSocket clients. Exploitation hinges on first capturing a signed request (for example from a TLS-terminating proxy's logs), which is why attack complexity is rated High under CVSS 5.9; no public exploit code was identified at time of analysis and there is no KEV entry.
Unauthenticated server-side request forgery (SSRF) in the WordPress plugin 'Generate PDF using Contact Form 7' before version 4.2.2 lets remote attackers force the server to fetch an arbitrary URL - including internal-only resources - and read the fetched content back out of the generated PDF. Exploitation requires only that the site runs the vulnerable plugin and exposes a Contact Form 7 form whose submission triggers PDF generation; no credentials or user interaction are needed (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C, assessed C:H versus the published C:L). A publicly available exploit exists (WPScan), and a vendor fix is available in 4.2.2, so upgrading is the decisive action.
Uncontrolled resource allocation in the VES Collector component of O-RAN-SC SMO OAM (2025-06-10) lets remote, unauthenticated attackers exhaust resources and degrade availability, with no confidentiality or integrity impact (VC:N/VI:N/VA:L). The CVSS 4.0 base of 6.9 reflects network reachability with no privileges or user interaction (AV:N/AC:L/AT:N/PR:N/UI:N) and a proof-of-concept exploit is publicly available via a GitHub gist, though there is no confirmed actively exploited (CISA KEV) status. The upstream project was notified through a bug report (LF-O-RAN-SC Jira SMO-204) but has not responded, so no vendor-released patch is identified at time of analysis; practical exposure is limited to deployments where the VES Collector is reachable on an internal telecom OAM/management plane.
Autobahn Python releases prior to 26.7.1 enforce the maxMessagePayloadSize receive limit against the compressed on-the-wire frame length rather than the inflated message size, so an unauthenticated remote client can send a small permessage-deflate frame that decompresses far past the configured application message cap and is nonetheless allocated, reassembled, validated, and handed to application callbacks. The impact is resource-exhaustion pressure scored at CVSS 5.3 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L, availability-only), and the advisory does not establish confidentiality or integrity impact despite the general 'Information Disclosure' tag; no public exploit identified at time of analysis. Exposure is limited to WebSocket endpoints that negotiate permessage-compress (permessage-deflate or the bzip2/snappy/brotli backends) - endpoints that do not accept a compression extension for the connecting client are unaffected.
Bypass of the registration CAPTCHA in the WordPress Clean Login plugin before 1.19 allows unauthenticated attackers to create accounts without solving the challenge when the stored session value is empty. Publicly available exploit code exists, but no confirmed active exploitation (CISA KEV) was reported, and our independent assessment rates this as a genuine low-to-moderate priority issue because the impact is limited to spam and automation abuse rather than compromise of existing accounts or the server. The vendor has released version 1.19 as the fixed release.
The Clean Login WordPress plugin before 1.19 lets unauthenticated attackers create user accounts on sites where registration is disabled, bypassing an administrative control intended to prevent open sign-ups. The registration handler fails to check the site's registration-enabled setting, so even locked-down sites become vulnerable; impact is limited to unauthorized low-privilege account creation (CVSS I:L) with no direct confidentiality or availability loss. Publicly available exploit code exists, and the vendor has released a patched version (1.19).
Unauthenticated visitors can strip the login- and URL-hiding protection from WordPress sites running the Hide My WP Ghost plugin before 7.0.11 by triggering the plugin's own loopback security-check request with a missing or incorrect verification value, causing the concealment to fail open and re-exposing the hidden WordPress login page location. The flaw is network-reachable, low complexity, and requires no authentication or user interaction (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N, base 5.3), but its impact is deliberately narrow: it only reverts the site to its ordinary, unhidden login posture and does not bypass WordPress authentication, so real-world risk depends on the strength of the underlying credentials and any additional login protections in place. Publicly available exploit code exists and the issue is tracked by WPScan as well as ENISA EUVD-2026-82686; a vendor fix is available in version 7.0.11, and there is no indication of confirmed active exploitation (no CISA KEV listing) and no EPSS score was provided in the source data.
Unauthenticated attackers can strip the hardening protections from the Hide My WP Ghost WordPress plugin (versions before 7.0.11) by supplying a single WooCommerce-looking request parameter, which the plugin accepts as proof of a legitimate WooCommerce request and uses to switch off its firewall, threat detection and login/URL-hiding logic - re-exposing the concealed login and admin URLs on any request. The issue affects default installations where the plugin's protection features are enabled (its core purpose), requires no authentication, user interaction or non-default configuration (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N, 5.3), and publicly available exploit code exists, with no confirmed active exploitation reported. The impact is indirect - WordPress core authentication is not bypassed - but any site treating this plugin as a meaningful obscurity control effectively loses that protection until it is upgraded to 7.0.11 or later.
Unauthenticated remote attackers can inflate a plugin option in the Really Simple Security WordPress plugin (versions 8.1.6 up to but not including 9.8.3) by supplying arbitrary address values that the plugin uses as storage keys without validation, causing unbounded option growth and slower handling of missing pages. The impact is availability-only degradation (CVSS 5.3, AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L) with no confidentiality or integrity effect, and default configurations are reachable by network unauthenticated exploitation as long as the site is online. Publicly available exploit code exists and WPScan published the finding, but there is no indication of confirmed active exploitation (CISA KEV); sites running 9.8.3 or later are unaffected.
Unauthenticated attackers can overwrite arbitrary appointment records in the Easy Appointments WordPress plugin (versions 4.0 through 4.0.2.2) by sending a request-supplied appointment id to the plugin's public appointment-reservation endpoint, which performs no ownership or authorization check before updating the record. The same missing check lets a follow-on cleanup routine delete the targeted appointments, so booking data on affected sites can be corrupted or removed. Publicly available exploit code exists, but there is no indication of confirmed active exploitation (the CVE is not in CISA KEV), and the impact is bounded - integrity and some availability of appointment records only, with no confidentiality loss and no code execution, which is why the risk is real but moderate rather than critical.
Unauthenticated attackers can add arbitrary products to a WooCommerce cart and check out at the discounted price tied to a Frequently Bought Together campaign in the UpsellWP WordPress plugin (versions 1.4.4 through before 2.2.10), because the plugin never verifies that cart items actually belong to the campaign being applied. The flaw is a business-logic validation gap rather than a data-exposure or availability issue: the manipulation is limited to an integrity impact (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N, base 5.3), with real-world loss scaling to the size of the configured discount and the value of the goods sold. Publicly available exploit code exists, but there is no indication of confirmed active exploitation; the vendor has released a fix, so upgrading to 2.2.10 removes the issue.
Unauthenticated attackers can create WordPress user accounts on sites running the Easy Form Builder by WhiteStudio plugin in versions 4.0.0 through before 4.2.0, even when the site owner has deliberately disabled user registration. The plugin fails to validate a submitted value against its stored configuration for certain form types (CWE-284, improper access control), so the registration-disabled setting is bypassed via a remote, unauthenticated request requiring no user interaction (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N, 5.3). Publicly available exploit code exists for this issue, but it is not confirmed as actively exploited (not in CISA KEV); the realistic impact is bounded by the site's default new-user role, typically Subscriber, and updating to 4.2.0 fully removes the flaw.
Process Compose before 1.120.0 exposes its MCP SSE listener on loopback without validating the Host or Origin header or authenticating the caller, allowing a malicious website to reach that listener through DNS rebinding and issue MCP requests. Unauthenticated remote attackers (CVSS 4.0 PR:N, UI:P) can, when expose_control_tools is enabled, enumerate process state, read, search or truncate logs, and start, stop, restart or scale local processes; user-defined tools may expose further commands and output. Publicly available exploit code exists (SSVC Exploitation: poc), no active exploitation is confirmed (no CISA KEV listing), and the issue is fixed in 1.120.0.
Unauthenticated attackers can cancel or confirm arbitrary appointments in the Easy Appointments WordPress plugin before 4.0.2.2 because the mail-link action token is derived from a hardcoded, source-embedded salt combined with the appointment's creation timestamp instead of an unguessable secret. Exploitation hinges on knowing or guessing the target appointment timestamp, which keeps reliability limited (reflected as AC:H in CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:L, base 4.8); impact is confined to integrity and availability of existing appointment records with no data disclosure. Publicly available exploit code exists, and a vendor-released patch (4.0.2.2) is available.
Unauthenticated information disclosure in CordysCRM (1Panel-dev) versions prior to 1.7.2 lets any network-reachable caller retrieve CRM form metadata via GET /mcp/form/config/{formKey}, exposing field names, types, required flags, default values, options, validation rules, and binding sources. The flaw exists because ShiroFilter marks the /mcp/** path as anonymous and McpController.getMcpField carries no permission annotation, so the endpoint is reachable without credentials (CVSS 4.0 base 6.9; independent assessment CVSS 3.1 AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N). Impact is limited to schema/metadata disclosure rather than direct access to customer records, and no public exploit code has been identified at time of analysis; exploitation does require knowledge or brute-forcing of valid formKey identifiers.
Missing authentication in the FastAPI backend of Forget-C Jellyfish AI Short Drama Studio (versions 0.1.0-alpha, 0.2.0, 0.3.0, 0.3.1 and 0.3.2) lets unauthenticated remote attackers reach API endpoints that should require credentials, because the dependency logic in backend/app/dependencies.py does not enforce an authentication check. The CVSS 4.0 base score is 6.9, and the vector (AV:N/AC:L/AT:N/PR:N/UI:N, LOW confidentiality/integrity/availability) indicates no credentials, no user interaction and no special conditions are needed; no public exploit code is identified at time of analysis and there is no confirmed actively exploited (CISA KEV) status. Real-world priority remains low because the software is an obscure self-hosted hobby project, so exposure depends on someone operating a publicly reachable instance, and a successful bypass yields only LOW impact across all three security objectives.
Unauthenticated remote attackers can query the SNMP v2c service on Grandstream GWN7660ELR access points running firmware before 1.0.27.6 when the service is left with the factory-default community string 'public', harvesting system metadata such as OS and kernel versions, process names and command-line arguments, interface configuration, routing and ARP tables, active TCP connections, and filesystem paths. The CVSS 4.0 base score is 6.9 (vector CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:N/VA:N/SC:N/SI:N/SA:N) and our independent assessment maps it to CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N - a read-only reconnaissance issue requiring only UDP/161 reachability, with no write, code execution, or lateral-movement primitive on its own. No public exploit code has been identified at time of analysis and the CVE is not in CISA KEV, so this is a genuine but modest-priority information-disclosure weakness whose real-world exposure depends on whether SNMP is enabled, still using the default community string, and reachable from untrusted networks; it is fixed in firmware 1.0.27.6.
Heap memory corruption in snappy-java 1.1.10.8 and earlier allows remote attackers to write attacker-influenced bytes past the bounds of the output array when an application decompresses untrusted compressed data through the typed Snappy.uncompress*Array APIs (such as uncompressIntArray, uncompressFloatArray, or uncompressDoubleArray), which size their result array using the uncompressed length divided by element size while forwarding the undivided length to the native JNI layer. Per the assessed vector (CVSS 3.1 AV:N/AC:L/PR:N/UI:N, C:N/I:L/A:L) no authentication or user interaction is needed, but exploitation is conditional: the consuming application must expose the typed decompression path to attacker-controlled input, and applications that only use the generic byte[] uncompress path are not exposed. Impact is currently rated Low integrity and Low availability (heap corruption and crash) with no confirmed confidentiality loss or demonstrated code execution, and no public exploit was identified at time of analysis; this is a genuine memory-corruption flaw but a moderate, not top-tier, priority, and no vendor-released fixed version is identified in the available data.
Improper UTF-8 percent-decoding in uri-js through 4.4.1 lets remote, unauthenticated attackers feed crafted percent-encoded URIs that the library converts into ASCII metacharacters such as '/', '\', '.' and CRLF, enabling downstream path traversal or CRLF injection wherever the decoded output is consumed without re-validation. The flaw is remotely triggerable with no authentication or user interaction (CVSS:4.0 base 6.9; independent assessment CVSS:3.1 AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N), but real-world harm is constrained by the consuming application: no code execution, no availability impact, and no effect at all if the downstream code re-validates the decoded string. There is no CISA KEV listing and no public exploit identified at time of analysis, so this is currently a latent input-neutralization weakness for applications that treat uri-js output as already-sanitized rather than a live exploitation campaign.
Heap-based out-of-bounds write in the PJSIP pjmedia AVI player (pjproject 2.17 and earlier) lets a crafted AVI file overwrite memory beyond a frame buffer when an application plays the file or pulls its frames. The buffer capacity is derived from the video dimensions declared in the file, while the copy length is taken from the file's own chunk header, so a malicious header produces an attacker-controlled write; the protection that would normally catch this is a pj_assert() that is compiled out of production release builds. The issue is local and file-driven, requiring the victim to open the media (CVSS v3.1 AV:L/UI:R, PR:N per the assessed vector), with crash-level impact for ordinary local playback and full memory corruption for applications that ingest untrusted AVI from relays, transcoders or user uploads; no public exploit code has been identified at time of analysis, no CISA KEV listing applies, and no fixed release version is available as of this review.
Bransys ELD ships with hardcoded MQTT credentials that allow unauthenticated remote attackers to subscribe to the real-time telematics feed for every active device belonging to the subset of carriers that were connected to the affected MQTT broker. The assessed CVSS:3.1 vector (AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N) confirms the weakness is remotely reachable, low complexity, and requires no authentication or user interaction, but confines impact to low confidentiality - read-only exposure of real-time data with no integrity or availability effect, consistent with the vendor-side CVSS 4.0 score of 6.9 (VC:L, VI:N, VA:N). No public exploit code has been identified at time of analysis, though the hardcoded-credential nature of the flaw means anyone who extracts the static string from firmware or a decompiled mobile/app binary can reuse it without further tooling; practical exposure is bounded by whether the broker stays network-reachable and still accepts the unchanged credentials.
Sandbox escape in vm2 through 3.12.0 lets untrusted guest code cross the NodeVM isolation boundary and change process-wide state by calling crypto.setFips(), which toggles the FIPS mode of the entire host Node.js process (observed as crypto.getFips() flipping from 0 to 1 in the reporter's test). The vulnerability is only reachable when the embedding application runs vm2 as a NodeVM with the non-default, opt-in configuration require.builtin: ['crypto']; authentication and user interaction are not required in that scenario, and the practical impact is limited to low-integrity and low-availability effects rather than arbitrary host code execution or data disclosure. No public exploit identified at time of analysis; the issue is fixed in vm2 3.12.1, available from the vendor (GHSA-x3v6-43hc-82mc).
An out-of-bounds read in GPAC's RTP depacketizer lets unauthenticated remote attackers crash or partially disclose memory by delivering a crafted 3GPP timed-text (TTXT) RTP payload with an oversized advertised length. The flaw lives in gf_rtp_parse_ttxt() in src/ietf/rtp_depacketizer.c and affects GPAC 26.08-DEV and earlier builds prior to abi-16.26; exploitation requires the target to be acting as an RTP receiver/depacketizer rather than simply parsing a local file. No public exploit code was identified at time of analysis and there is no confirmed active exploitation (CISA KEV), so this is a genuine but moderate-severity memory-safety defect whose practical reach depends on GPAC being deployed in an RTP-consuming configuration.
CordysCRM 1.7.3 and earlier allows an authenticated user holding the PROCESS_SETTING_ADD permission to store an attacker-chosen webhook URL in an approval-flow node and then cause the server to issue outbound HTTP requests to that URL when an approval action is approved, turning the application into a proxy for reaching internal-only services. The assessment rates this CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:N/A:N - high confidentiality impact against the attacked component with no integrity or availability impact - and the practical payoff is cloud metadata access (for example 169.254.169.254) and internal network reconnaissance, bounded by the server's network reach and by the multi-step trigger (configure via POST /approval-flow/add, fire via POST /approval-action/approve). The flaw is fixed in 1.7.4 per vendor advisory GHSA-fg6q-pfj7-fghw, PR #2966 and commit e0ae23e; no public exploit code and no confirmed active exploitation were identified at time of analysis, and exploitation requires a low-privileged authenticated account plus a second in-application approval step.
FluidSynth builds compiled with the native DLS parser enabled can be driven into a local denial of service by a crafted DLS SoundFont file: an integer overflow in the parser's articulation-chunk size check lets the loader run roughly a billion out-of-bounds 12-byte iterations, causing a crash or hang. The flaw affects versions 2.5.0 through 2.5.5 and is fixed in 2.5.6; exploitation requires the target to be built with the non-default CMake option enable-native-dls set to ON and to load the attacker-supplied file (unauthenticated per the CVSS PR:N vector, but the attack vector is local and the malicious file must reach the application). No public exploit code or confirmed active exploitation was identified at time of analysis, and impact is confined to availability - there is no code execution or data disclosure.
A 32-bit integer overflow in FluidSynth's native DLS parser lets a crafted DLS sound-bank file bypass the ptbl chunk-size check, triggering an approximately 4 GB pool-cue allocation and billions of out-of-bounds reads that crash the process, producing a denial of service against the affected 2.5.0 through 2.5.5 releases. Exploitation requires the library to be built with enable-native-dls=ON and an attacker-supplied DLS file to be opened by the application; the CVSS vector indicates no privileges are required (PR:N), and no public exploit code was identified at time of analysis. The vendor has fixed the flaw in FluidSynth 2.5.6.
Denial of service in WinFsp through 2.2.26215 allows an authenticated, unprivileged local user to crash the Windows kernel by issuing FSP_IOCTL_TRANSACT requests to the WinFsp control device, triggering a null pointer dereference in the driver's Fast I/O device control handler. The vulnerability requires local access (PR:L) and a system with WinFsp installed; it results only in a system crash with no data exposure or code execution. No public exploit code has been identified at time of analysis, and the flaw is not known to be actively exploited.
Privileged-mode password recovery affects PLANET IGS-5225-8P2T4S industrial managed switches running V1 firmware before 1.2412b260707 or V2 firmware before 2.2412b260519, because these devices store administrative credentials using the weak MD5 hashing algorithm. An attacker who obtains the device configuration file - typically via an admin config export, an exposed backup, or physical/local device access - can crack the stored hash offline and recover the privileged-mode password. Exploitation is not remote and requires an authenticated, high-privileged foothold or local access to the config file (CVSS 3.1 AV:L/AC:L/PR:H/UI:N, confidentiality-only impact), and no public exploit has been identified at time of analysis; this is a genuine but low-urgency information-disclosure weakness rather than a critical remotely-reachable flaw.
Improper exception handling in pg_partman's run_maintenance() function allows an authenticated database role holding partman_user access to permanently stall automated partition maintenance across an entire database. Because the exception handler wraps the whole row-processing loop over part_config rather than each individual partition set, a single crafted row that reliably raises an error - and that is given a low maintenance_order value so it is processed first - immediately aborts the loop and skips every legitimate partition set scheduled after it. Each subsequent maintenance tick (background worker or cron) fails the same way, so child partitions stop being created and retention stops dropping old ones, producing a database-wide availability impact; no public exploit code has been identified at time of analysis and there is no KEV listing, though EPSS and exploitation data were not available in this feed. Fixes are available in pg_partman 5.5.0.
Unvalidated OIDC redirect handling in the Kyoo self-hosted media server before 5.1.0 lets an unauthenticated attacker capture a victim's one-use login token and exchange it for that victim's Kyoo session. The attacker starts the OIDC login flow with an attacker-controlled redirectUrl; because the pre-fix code stored that URL alongside the opaque login state and appended the token, provider, and error values to it without validating the destination or binding the token to the browser session that began the flow, a social-engineered victim who completes authentication at their identity provider has the token delivered to the attacker's host, where it can be replayed through /auth/oidc/callback/{provider}. CVSS 3.1 base score is 6.5 (AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:N/A:N) - network reachable and requiring no privileges or authentication, but gated on victim user interaction and on the instance having at least one OIDC provider configured and enabled; no public exploit identified at time of analysis.
HTTP request smuggling in Netty's HTTP/1.1 decoder (HttpObjectDecoder and the HttpRequestDecoder built on it) lets remote, unauthenticated attackers send a chunk-size line containing an embedded bare carriage return - for example `0\rX` - because strict chunk-line validation is skipped whenever the size line carries no `;` chunk extension, causing a request-boundary desync when Netty shares a connection with a second HTTP intermediary that frames the message differently. The flaw reaches every application and Red Hat product that embeds Netty, including AMQ Broker 7, AMQ Clients, Quarkus, Camel 4 for Quarkus 3, Camel for Spring Boot 4, Keycloak / Red Hat Single Sign-On, JBoss EAP 7 and 8, Fuse 7, Data Grid 8, Apicurio Registry 3 and Debezium 3, while independent assessment keeps it at CVSS 6.5 with AC:H because a standalone Netty service with no disagreeing proxy in the chain yields little practical smuggling impact. The public report includes a reproducible PoC test harness, but there is no CISA KEV listing and no evidence of active exploitation.
IBM QRadar deployments running 7.5.0 through 7.5.0 UP15 Interim Fix 006 expose sensitive backup-file contents to any holder of a valid QRadar account, because those backup files are stored with incorrect permissions. Exploitation requires authentication (CVSS vector PR:L) but no administrative rights and no user interaction, and an attacker who satisfies that bar can read configuration, credential, or event data captured in the affected backups, yielding a high confidentiality impact (C:H) with no integrity or availability effect. This is a moderate-priority information-disclosure issue rather than a high-urgency emergency: no public exploit code has been identified at time of analysis, the flaw is not in CISA KEV, and the authentication prerequisite plus the need for an unpatched interim-fix level are the principal limiting factors.
Adjacent-network attackers can drive the Name Service component of IBM WebSphere Application Server 8.5 and 9.0 into a denial-of-service condition by supplying maliciously crafted serialized data to the CORBA/IIOP naming endpoint, which the JVM then deserializes. The flaw is unauthenticated and requires no user interaction (CVSS:3.1/AV:A/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, 6.5), so the practical barrier is network position rather than credentials: the attacker must be able to reach the Name Service on the same or an adjacent network segment, and default configurations leave that service listening on the network. Impact is confined to availability - no loss of confidentiality or integrity is assessed - and no public exploit code or confirmed active exploitation has been identified at time of analysis; IBM has published a vendor fix.
HTTP request smuggling in Netty's netty-codec-http component lets unauthenticated remote attackers bypass front-end security controls by sending a crafted HTTP/1.1 chunk-size token containing whitespace after the hexadecimal digits, which Netty parses differently from other HTTP intermediaries. The flaw carries a CVSS 3.1 base score of 6.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:L), meaning no authentication or user interaction is required, but integrity and availability impact is limited; meaningful exploitation generally depends on a multi-tier topology where a proxy, load balancer, or backend parses the chunk-size token inconsistently with Netty. Red Hat ships the affected library across a broad portfolio - including AMQ Broker 7, AMQ Clients, Quarkus, Keycloak and Red Hat Single Sign-On 7, JBoss EAP 7/8, Fuse 7, Data Grid 8, Camel for Spring Boot 4, Camel 4 for Quarkus 3, Apicurio Registry 3 and Debezium 3 - and no public exploit code has been identified at time of analysis, with no confirmed active exploitation (CISA KEV).
Server-side request forgery in OpenCVE versions prior to 3.0.0 lets an authenticated user who holds permission to configure notification channels force the OpenCVE server to issue HTTP or HTTPS requests to arbitrary attacker-chosen destinations, including localhost interfaces, RFC1918 internal ranges, link-local addresses and the cloud metadata endpoint 169.254.169.254, with the response content retrievable by the attacker. The prerequisite is the notification-configuration privilege (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N, CWE-918), so anonymous actors cannot trigger it, and the practical impact is highest in cloud-hosted deployments where an instance metadata service is reachable from the OpenCVE host; no public exploit code has been identified at time of analysis. The issue is fixed in OpenCVE 3.0.0.
Stack exhaustion in OpenImageIO's FITS image reader (all releases prior to 3.0.20.0, 3.1.15.0, and 3.2.0.3-beta1) lets a crafted FITS stream crash a parsing process by driving FitsInput::read_fits_header() into unbounded recursion. The trigger is a stream of consecutive 2880-byte header blocks that never contain the mandatory END keyword, so each block causes another self-call until the application stack is consumed; remote, unauthenticated attackers can deliver the payload but successful exploitation requires the victim to open or ingest the file (UI:R), and the resulting impact is availability-only with no code execution or data disclosure. No public exploit code or confirmed active exploitation has been identified at time of analysis, and vendor-released fixes are available.
Virtual host isolation bypass in IBM WebSphere Application Server and WebSphere Application Server Liberty, as bundled in IBM CICS TX Advanced 10.1, allows remote requests addressed to one configured virtual host to reach resources belonging to another virtual host on the same instance. The vendor assigns CVSS 3.1 6.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N, CWE-284); our independent assessment keeps the same unauthenticated, low-confidentiality/low-integrity, no-availability profile but raises attack complexity to AC:H because the weakness is only meaningful when an instance is configured to serve multiple distinct virtual hosts, while a default single-virtual-host deployment offers little to bypass. No public exploit code and no confirmed active exploitation (CISA KEV) were identified at time of analysis, and a vendor patch is available per the IBM advisory.
HTTP request smuggling in IBM WebSphere Application Server 8.5 can allow unauthenticated remote attackers to desynchronize request processing when the server sits behind or in front of an HTTP intermediary with mismatched Content-Length parsing, potentially leading to credential theft, cache poisoning, or request hijacking. The flaw (CWE-444) rates 6.5 (CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:L/A:N) and requires a specific proxy-layer mismatch and concurrent traffic; no public exploit identified at time of analysis. Vendor patch is available and the risk is assessed as moderate and non-urgent.