Improper TLS certificate validation in TikTok Video Downloader PHP allows unauthenticated remote attackers who already hold a man-in-the-middle position to supply forged HTTPS certificates to index.php cURL requests, compromising the integrity of data fetched from TikTok APIs and CDN endpoints. The issue affects rolling-release revisions up to commit c4230497bb23dd9c1f81456fe8a982d951feac0d and is fixed by commit 64c0155abb88e714198241d41b918cefc0d5a42a; exploitation requires a MITM position, and no public exploit code was identified at time of analysis. Impact is limited to integrity (CVSS 6.3, high attack complexity) with no confidentiality or availability effect.
Improper TLS certificate validation in GlitterWare Passy through 1.10.1 lets an unauthenticated attacker who already holds an on-path network position impersonate HTTPS hosts and intercept or alter traffic the app exchanges. Exploitation is rated high complexity because the attacker must actively man-in-the-middle a connection at the moment Passy connects, but no user interaction is required; no public exploit was identified at time of analysis. Upgrading to 1.10.2 removes the insecure global HTTP override and restores certificate validation.
Improper TLS certificate validation in fatfree-core through 3.9.2 leaves outbound HTTPS requests made through the framework's HTTP client open to man-in-the-middle attacks. The flaw hard-disables CURLOPT_SSL_VERIFYPEER in web.php's _curl function, but exploitation is difficult: it requires the application to use this client for outbound HTTPS and an attacker to hold an active MITM position, and impact is limited to data transmitted through that client. The vendor released 3.9.3 with commit a2c755c, and no public exploit code or active exploitation is reported; this is a medium-severity, low-urgency issue rather than an emergency.
Local information disclosure in Dell Application Installer before 26.09.0.1 exposes sensitive data that the product stores without encryption, allowing any local user on the host to read it without authentication (CVSS 6.2, PR:N). The flaw requires local access to the affected machine but no privileges or user interaction, so shared, multi-user, or physically accessible Dell systems are the primary exposure surface. No public exploit code or active exploitation has been identified at time of analysis, and Dell has released version 26.09.0.1 to remediate the issue.
Stored/DOM cross-site scripting in Mercury's mr.Markdown widget (versions 3.0.0 through 3.2.10) allows attacker-supplied Markdown containing raw HTML event handlers or javascript: links to execute in the browser of any user who views the affected dashboard. The widget converts Markdown with python-markdown, which passes raw HTML through, and then renders it into an ipywidgets.HTML sink via innerHTML without sanitization, so untrusted database rows, form submissions, uploaded files, or LLM/API responses can become script execution; the prior mr.Message markdown path uses the same unsafe renderer. Exploitation is unauthenticated per the CVSS vector (PR:N) but requires a victim to view the application (UI:R), and no public exploit code was identified at time of analysis. Mercury 3.2.11 fixes the issue by sanitizing rendered Markdown by default and requiring an explicit unsafe_allow_html=True opt-in for trusted content.
A permission-check flaw in Zephyr's RTIO syscall layer lets an unprivileged user-mode thread write into memory it is only permitted to read. In z_vrfy_rtio_cqe_get_mempool_buffer(), the two output parameters buff and buff_len are validated with K_SYSCALL_MEMORY_READ() instead of K_SYSCALL_MEMORY_WRITE(), so when CONFIG_USERSPACE and CONFIG_RTIO_SYS_MEM_BLOCKS are both enabled and the thread already holds an RTIO context grant, the kernel performs a pointer-sized and a 4-byte supervisor-mode store (a CWE-787 out-of-bounds write through mis-validated output pointers). Pointing those pointers at global read-only flash (.text/.rodata) causes a kernel fault handled by z_fatal_error() and the default fatal handler halts or reboots the device - a reliable denial of service - while targeting a K_MEM_PARTITION_P_RW_U_RO partition yields a partly attacker-controlled write that bypasses user-read-only protection; no data disclosure is possible because the read check still applies, and there is no public exploit identified at time of analysis, with exploitation strictly local (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:L/A:H, 6.1).
Cross-site scripting in the IBM Concert Web UI (versions 1.0 through 3.0.0) lets remote attackers inject arbitrary JavaScript into pages viewed by other users, potentially disclosing credentials from within a trusted session. Exploitation requires the victim to interact with attacker-supplied content - such as clicking a crafted link or loading a page reflecting the payload - while authenticated to Concert, so practical risk is targeted rather than mass exploitation. No public exploit code or confirmed active exploitation is known, and IBM has released a vendor fix; EPSS context is not provided in the available data.
Login cross-site request forgery (CSRF) and open redirect in the Apache Camel OAuth component (camel-oauth) let an attacker silently sign a victim into an attacker-controlled identity: the authorization code callback in affected releases neither issued a session-bound state parameter nor validated one, so a code obtained in any browser can be redeemed against whatever session presents it. Affected are Apache Camel 4.11.0 before 4.18.5 and 4.19.0 before 4.22.1 - and only deployments that actually use the camel-oauth authorization code flow to authenticate users, since that component is optional. The CVSS 3.1 score is 6.1 (AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:N): remote and unauthenticated from the attacker's side, but exploitation requires victim interaction (following a crafted callback link), and impact is confined to the OAuth login identity rather than the host. No public exploit code or confirmed active exploitation was identified at time of analysis; the vendor has released fixed versions (4.23.0, 4.18.5, 4.22.1).
Reflected cross-site scripting in Ellucian Banner Student Registration Self-Service (StudentRegistrationSsb) 9.23 and earlier allows a remote, unauthenticated attacker (CVSS 3.1 AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:L/A:N, base 6.1) to run arbitrary script in a victim's browser session by tricking the victim into opening a crafted link to the /ssb/registration endpoint whose mepCode parameter carries the payload. The flaw is gated by user interaction - the attacker cannot trigger execution without a victim clicking the link - so it is best suited to phishing or targeted social engineering against students and registration staff rather than self-propagation. There is no confirmed active exploitation (not listed in CISA KEV) and no public exploit identified at time of analysis, but the absence of a vendor-supplied fix for all versions up to and including 9.23 means defenders must rely on WAF, output-encoding and network-exposure controls rather than patching.
Local privilege escalation in Red Hat rhc's rhc-canonical-facts.service lets an attacker who has already compromised the privileged yggdrasil service account pre-plant a symbolic link so the root-run unit follows it and changes ownership of an arbitrary file, enabling escalation to root. The flaw is CWE-59 link following and affects rhc installations on Red Hat Enterprise Linux 7, 8, 9, and 10 per CPE data; exploitation is local-only and requires high privileges (CVSS PR:H), so risk is confined to systems where the yggdrasil account is already reachable. No public exploit code was identified in the provided sources at time of analysis, and no CISA KEV or EPSS signal was supplied.
Denial of service in the Zephyr UpdateHub firmware-over-the-air client can be triggered by a malformed CoAP probe response that omits the sha256sum field, causing z_impl_updatehub_probe() to call strlen() on a NULL pointer. The issue affects Zephyr builds with UpdateHub enabled, especially the default plain-CoAP configuration without CONFIG_UPDATEHUB_DTLS, where an unauthenticated on-path, DNS-spoofing, or source-address-racing attacker can inject the response during automatic polling. Impact is availability only and target-dependent: systems that protect or leave unmapped address 0 halt in Zephyr's fatal handler, while classic Cortex-M targets with a readable vector table may merely reject the request; no public exploit has been identified at time of analysis.
Denial of service in IBM MQ client applications that consume messages asynchronously can be triggered when a queue name carrying an improperly validated length value reaches the client, producing a buffer overflow (CWE-120) that crashes or hangs the consuming application. The flaw affects IBM MQ Operator and IBM MQ Advanced Container Images deployments and is reachable either by an attacker who already controls the queue manager or by a man-in-the-middle able to tamper with message metadata on unencrypted, non-TLS channels. IBM has released a vendor patch per its support advisory, no public exploit code or active exploitation has been identified at time of analysis, and the CVSS 5.9 score reflects availability-only impact with high attack complexity.
IBM Cloud Pak System 2.3.6.0 and 2.3.6.1 employ weaker-than-expected cryptographic algorithms, allowing an attacker who obtains the protected ciphertext to decrypt highly sensitive information. The CVSS 5.9 score (AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N) reflects that exploitation requires capturing the encrypted data and performing cryptanalysis, which substantially constrains real-world risk despite the unauthenticated network vector. No public exploit code or active exploitation has been identified at time of analysis, and IBM has released a vendor patch.
Signature verification in the Apache Camel PQC component can be silently defeated when concurrent exchanges share the producer's single stateful java.security.Signature instance, allowing a message to be reported as verified even though its signature was never actually checked against its own payload. Affected are Apache Camel releases from 4.12.0 before 4.18.5 and from 4.19.0 before 4.22.1 that use the camel-pqc sign/verify operations; remote, unauthenticated attackers who can drive concurrent traffic through such a route may interleave init/update/verify sequences. Exploitation is timing-dependent and non-deterministic (CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:H/A:N, 5.9), and no public exploit was identified at time of analysis; the practical prerequisite is a concurrently invoked PQC signing endpoint, which is the ordinary condition for a loaded Camel route. Vendor-released fixes exist in 4.23.0, 4.18.5 and 4.22.1.
Man-in-the-middle interception of Apache Camel's Knative HTTP client can expose traffic when TLS is enabled without a configured truststore, because affected versions (4.0.0 before 4.18.5 and 4.19.0 before 4.22.1) fall back to trusting every presented certificate instead of the JVM default trust anchors. Exploitation is conditional rather than universal: the deployment must set camel.knative.client.ssl.enabled=true while leaving both truststore.path and trust.cert.path unset, and a remote unauthenticated attacker must already hold a privileged network position to terminate the TLS session (CVSS 3.1 5.9, AC:H, C:H/I:N). No public exploit identified at time of analysis, and the vendor fix in 4.23.0 (plus 4.18.5 and 4.22.1 LTS) leaves trust options unset so JVM defaults apply and requires an explicit trust.all property to opt into trust-all.
In the camel-thrift data format, unmarshalled messages are deserialized into a process-wide shared defaultInstance and that same object is returned as the exchange body, so state from one message survives into the next. Senders who omit optional fields in a later payload inherit the previous sender's values, and concurrent exchanges can observe each other's partially written state - disclosing one tenant's data to another. Affected are Apache Camel 4.0.0 through 4.18.4 and 4.19.0 through 4.22.0; no public exploit identified at time of analysis, and exploitation requires a genuinely multi-tenant route using the Thrift data format plus attacker influence over message sequencing or timing (CVSS 5.9, AC:H, PR:N).
Undetected tampering with PGP-encrypted messages is possible in the Apache Camel Crypto PGP data format because unmarshal() only validated the OpenPGP modification detection code when the incoming message used the integrity-protected packet type, and the packet type is selected by the message producer. Deployments running Camel 4.0.0 through 4.18.4 or 4.19.0 through 4.22.0 that decrypt externally supplied PGP ciphertext are affected; unauthenticated remote attackers who can produce those messages, or rewrite them in transit, can substitute the legacy symmetrically encrypted data packet and have malleated content accepted without any integrity failure. Impact is limited to integrity (silent forgery or modification of decrypted content) with no confidentiality or availability loss, so this is a genuine but moderate-severity flaw rather than an emergency; no public exploit code was identified at time of analysis and there is no confirmed active exploitation (CISA KEV) - exploitation requires the attacker to control or intercept the message bytes, which is why the vector carries AC:H.
Improper credential verification in the Apache Camel camel-shiro security processor lets a remote attacker whose request names an already-in-use principal ride a thread-bound Shiro subject and skip the password check entirely. Affected are Camel 4.0.0 through 4.18.4 and 4.19.0 through 4.22.0 that secure routes with ShiroSecurityPolicy, particularly when alwaysReauthenticate=false keeps subjects long-lived on pooled worker threads; the CVSS:3.1 vector is AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N (5.9), reflecting unauthenticated network reachability gated by high attack complexity. Because exploitation depends on thread-pool reuse timing and an exact principal-name match, it is non-deterministic rather than reliably reproducible, and no public exploit code or confirmed active exploitation (CISA KEV) was identified at time of analysis. Successful abuse yields access to protected routes and their data as the impersonated principal, with no integrity or availability impact assessed.
Apache Camel's Docling component implicitly interprets String message bodies by prefix, letting attacker-influenced content trigger server-side request forgery (bodies beginning with http:// or https://) or local file disclosure (bodies beginning with / or containing a backslash) without any endpoint option to pin the interpretation mode. Affected are Camel 4.15.0 through 4.18.4 and 4.19.0 through 4.22.0; the flaw is unauthenticated from the component's perspective (CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N, 5.9) but requires that a route actually forward externally influenced String bodies into a docling endpoint, and confidentiality-only impact is rated High. No public exploit code or CISA KEV entry is associated at time of analysis; EPSS data was not provided. Vendor-released patches exist (4.23.0, 4.22.1, 4.18.5) along with new opt-in hardening options.
Denial of service in the Zephyr RTOS Bluetooth Classic Hands-Free Profile (HF role) allows a remote Audio Gateway peer to permanently leak entries from the fixed HFP TX buffer pool and then deadlock the Bluetooth RX thread, hanging the device until reboot. The bug lives in subsys/bluetooth/host/classic/hfp_hf.c: hfp_hf_disconnected() cancels work items and clears hf->acl but never drains the AT-command buffers queued on hf->tx_pending, and it leaves BT_HFP_HF_FLAG_CONNECTED set, while hfp_hf_create() memsets a reused HF slot without checking whether any work item is still pending or running. Only builds with CONFIG_BT_HFP_HF enabled are affected, and the attacker must reach an established, authenticated HFP session (paired/SSP devices have L2CAP authentication enforced before the RFCOMM channel is reachable), so this is an adjacent-range, authenticated-session availability issue rather than a drive-by one. No public exploit code was identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV), but the per-cycle leak is trivially repeatable by any connected AG.
Improper authentication in the dotnet eShop Catalog API's AddApplicationServices configuration lets unauthenticated remote attackers bypass authentication controls and achieve limited confidentiality, integrity, and availability impact on catalog data; affected builds are up to commit 913acea66708a59595351174cee9b29f37179d29. The CVSS v4 vector records Proof-of-Concept exploit maturity (E:P), so publicly available exploit code exists, but there is no confirmed active exploitation (not in CISA KEV). Because eShop is a Microsoft reference/sample application rarely exposed in production, practical risk is moderate and bounded to low-impact catalog-level effects.
Heap buffer overflow in GPAC MP4Box's AC4 descriptor parser can crash the process when a local user opens a crafted AC4 file, producing a denial of service. Affected builds include the 2.5-DEV-rev1593-gfe88c3545-master development snapshot; the CVSS 3.1 vector is AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H, so exploitation requires victim interaction and affects availability only. EPSS is 0.22% (12th percentile), and no public exploit identified at time of analysis; an upstream fix commit is referenced.
A heap use-after-free in GPAC/MP4Box (build 2.5-DEV-rev1593-gfe88c3545-master) lets a local attacker crash the tool by getting a victim to open a crafted AC4 audio file, producing a denial of service with no confidentiality or integrity impact. The flaw sits in the AC4 configuration cleanup path (gf_odf_ac4_cfg_clean_list in /odf/descriptors.c) and the surrounding AC4 substream-group and presentation parsing, which the upstream fix hardens with null checks, array bounds validation and de-duplication of temporary group lists before freeing. Scoring is CVSS 3.1 5.5 (AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H), EPSS is low at 0.17% (6th percentile) and there is no CISA KEV entry; a public PoC reference exists in the upstream GitHub issue and a linked social-media post, but it is not verified as working exploit code. The upstream fix commit e02d1fd24cdc26acb1b236ab38b3832cffcae21b is available, and no released patched version was independently confirmed.
GPAC MP4Box 2.4 contains a use-after-free in the filter pipeline that can crash the process when it processes a crafted media file. The CVSS vector requires local access and user interaction (AV:L/UI:R) with no privileges required (PR:N); impact is limited to availability (A:H), with no confidentiality or integrity effect. No public exploit was identified at time of analysis, and EPSS is low at 0.17% (6th percentile), consistent with a low-priority, high-noise issue rather than an emergency.
IBM Storage TS4500 Library versions 1.1.1.6 through 1.12.0.3 fail to invalidate a session after logout, allowing an authenticated user who obtains or retains a still-valid session identifier to impersonate another user. Exploitation requires a second user to have logged out and the attacker to possess that session token, so impact is limited to multi-user or insider scenarios and the issue is moderate severity/low urgency. Patch is available per IBM advisory node/7292613; no public exploit identified at time of analysis.
Denial of service in GIMP is triggered by importing a crafted SBZ palette file that lacks a color-model attribute, causing a NULL pointer dereference during string comparison in the palette parser. The flaw affects GIMP as packaged in Red Hat Enterprise Linux 6 through 10, requires local delivery of a malicious file and user interaction to import it, and results only in application crash with no confidentiality, integrity, or code-execution impact. No public exploit code or confirmed active exploitation was identified at time of analysis.
SQL injection in the Criminal Management System (vraj404) up to commit 2808d057e7ddc5295519a487365b0cacfd0ba8c4 lets remote unauthenticated attackers influence SQL queries through the home.php Submission Form parameters id, name, offName, crime, dob, arrDate, crimeDate, sex, address, folder and others, with limited confidentiality, integrity and availability impact. Publicly available exploit code exists, but no CISA KEV entry was provided, and the project uses a rolling release model with no fixed version or vendor response reported.
The FileBasedKeyLifecycleManager in Apache Camel's PQC (post-quantum cryptography) component writes generated private key material to disk using whatever permissions the process umask produces instead of restricting the file to its owner, so on hosts with a permissive umask such as 022 the unencrypted private key is group- and world-readable. Affected are Apache Camel 4.12.0 through 4.18.4 and 4.19.0 through 4.22.0 that use this component to persist key material; an attacker who already holds a separate local account on the same host can read the key file and then forge signatures or break key exchanges performed by that deployment. The CVSS 3.1 vector (AV:L/PR:L/C:H) confines this to a local, authenticated party - no public exploit code has been identified at time of analysis and there is no CISA KEV listing - and the vendor assesses severity as low because the remote sender that Camel's threat model targets is not the exposure path.
Apache Camel's camel-mllp component writes HL7 clinical message payloads containing protected health information into ordinary application logs, because the logPhi option defaulted to true and two additional code paths logged payload content even when an operator had explicitly set logPhi=false. Deployments running Camel 4.0.0 through 4.18.4, or 4.19.0 through 4.22.0, that use the MLLP transport for healthcare integration are exposed out of the box; anyone who can read the resulting logs - local authenticated users per the CVSS vector (AV:L/PR:L, C:H) - can recover patient data that was never meant to land in general-purpose log storage with its broader retention, aggregation and access surface. This is a genuine data-exposure defect but a low-urgency one in attacker terms, since it is not remotely triggerable and merely requires pre-existing log read access; no public exploit code was identified at time of analysis, and vendor patches are available. Note that simply setting logPhi=false is not a complete workaround, as two code paths continue to log payloads until the fix is applied.
Authenticated IBM MQ clients can cause a bounded denial of service and obtain limited information about process memory layout in IBM MQ Operator and IBM MQ Advanced Container Images deployments by sending a specially crafted request that triggers an out-of-bounds read. The issue requires a valid MQ client session and network access to the queue manager listener, and impact is limited to availability and low-confidentiality disclosure rather than code execution or data modification. No public exploit was identified at time of analysis, and IBM has published a vendor patch.
Refresh-token validation failure in the ForgeRock Key Manager component used by WSO2 API Manager, Universal Gateway, and API Control Plane allows a remote attacker who possesses a valid refresh token to present it as an access token and reach protected resources without completing the required client-credential exchange. The CVSS 3.1 score is 5.4 with AC:H because exploitation is gated by prior acquisition of a refresh token; no public exploit code and no CISA KEV entry are identified in the provided data.
HTML injection (CWE-80) in IBM Cloud Pak System 2.3.6.0 and 2.3.6.1 lets a remote attacker who already holds valid credentials on the web interface inject malicious HTML/script that executes in another user's browser within the hosting site's security context. Impact is limited to low confidentiality and integrity effects (CVSS 5.4) and requires both a legitimate authenticated account and a victim subsequently viewing the injected content, so it is a routine patch-management item rather than an urgent one. No public exploit code or confirmed active exploitation was identified in the available data; the vendor has published a fix advisory.
Missing authorization in the Surfer WordPress plugin through 1.8.0.744 lets an authenticated low-privileged user exploit an incorrectly configured access-control check to read or modify limited site data over the network without user interaction. The issue is rated CVSS 5.4 (AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:N), and no public exploit was identified at time of analysis.
Missing authorization in ThimPress LearnPress through 4.4.10 allows authenticated low-privilege users, such as self-registered students or subscribers, to perform actions they should not be permitted to perform against LMS-managed data. The flaw is medium severity (CVSS 5.4, CWE-862), remotely reachable with low complexity, and requires only a low-privilege authenticated account with no victim user interaction (PR:L/UI:N). No public exploit code or CISA KEV active exploitation is identified at time of analysis.
Information disclosure in IBM Concert 1.0 through 3.0.0 exposes schema suggestion data to network-reachable attackers; the CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N) indicates no authentication or user interaction is required, although the impact is limited to low confidentiality loss with no integrity or availability effect. The flaw is rated 5.3 (moderate) and is chiefly useful for reconnaissance rather than direct compromise. No public exploit code or CISA KEV entry was identified at time of analysis, and patch availability is documented in the vendor advisory.
Log injection in IBM Concert 1.0 through 3.0.0 lets remote unauthenticated attackers corrupt or forge log records by supplying input containing special or control characters that the product writes to its log files without neutralization. The CVSS 3.1 score is 5.3 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N), so the only assessed impact is limited integrity damage to log data - there is no code execution, data disclosure, or availability loss. No public exploit code or confirmed active exploitation was identified at time of analysis, and real-world severity depends on how downstream log consumers (viewers, SIEM parsers, audit reviews) handle unescaped content.
MISP before 2.5.48 can leak restricted tag metadata to a remote synchronization peer because its push-preparation routine removes non-exportable tags only after processing the entire report set, leaving all but the final report unsanitized. The issue affects authenticated MISP deployments that push events containing multiple event reports to another MISP server, and the same logic can also emit an empty report and cause the remote instance to fail saving the event. It is a low-severity confidentiality issue with no public exploit identified at time of analysis.
Authorization bypass in Cal.com and cal.diy through 6.2.0 allows any authenticated user to manipulate the ID argument in the webhook creation procedure and attach or alter webhooks on team-scoped event types or team webhooks without being an ADMIN or OWNER. The pre-patch code verified only eventType.userId, so team-owned resources were reachable by non-members; impact is limited to webhook configuration and secrets rather than full account takeover (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L). No public exploit was identified at time of analysis, and an upstream fix is proposed in PR #30223 but a released patched version was not independently confirmed.
Authorization bypass in sglang-router (sgl-model-gateway) through 0.3.2 allows an authenticated holder of a data-plane API key to reach the /ha/* mesh high-availability management routes that are meant to require the control-plane Admin role. Such a caller can read cluster topology, write gossiped configuration, modify global rate limits, and invoke POST /ha/shutdown to quietly dissolve HA coordination. Exploitation requires mesh/HA coordination mode to be deployed and a valid data-plane credential; no public exploit code has been identified at time of analysis, and a vendor patch is available.
Server-side request forgery in Sigma File Manager through 2.2.0 allows a malicious extension or poisoned catalog entry to point extension download commands at loopback, RFC1918, or cloud metadata endpoints, with the fetched response materializing as a file inside the extension directory rather than being streamed back to the attacker. The Tauri/Rust backend's validate_remote_url() in src-tauri/src/extensions/security.rs inspects the URL string but never checks the IP addresses the hostname resolves to, so a name resolving to 127.0.0.1 or 169.254.169.254 passes validation. Exploitation is conditional on the user installing or adding the attacker-supplied extension (CVSS 3.1 AV:N/AC:L/PR:N/UI:R, C:L only); no public exploit code identified at time of analysis and the issue is not in CISA KEV. Risk is highest when the app runs on a cloud instance exposing an instance-metadata endpoint.
Mass assignment in FlyCms through commit edfdf724 lets any authenticated low-privileged user POST extra User bean fields to /ucenter/account_update and persist protected account columns. An attacker can alter their own review status, rename the account, or set a victim's username, email, or mobile number to collide with an existing account and block that victim's login. The flaw is rated medium severity (CVSS 4.0 base 5.3, CWE-915); no public exploit code or confirmed active exploitation was identified at time of analysis.
Authentication bypass in the TP-Link Tapo C210 V3.60 DS JSON API allows attackers on the adjacent network to execute protected camera operations, including pan-tilt-zoom (PTZ) control, without a valid authenticated session. The flaw is rated moderate severity (CVSS 5.3; CVSS:3.1/AV:A/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N) because exploitation is confined to the same LAN/Wi-Fi segment and the impact is limited to camera functionality manipulation rather than full device takeover. No public exploit code was identified at time of analysis, and CISA KEV status does not indicate active exploitation; TP-Link has released patched firmware.
Username enumeration in WSO2 Identity Server allows unauthenticated remote attackers to distinguish valid accounts from invalid ones by observing inconsistent responses from the authentication endpoint when a supplied username does not exist. The issue affects deployments using the Email OTP Authenticator and WSO2 Carbon Abstract OTP Authenticator, and it is a low-to-moderate severity information-disclosure flaw (CVSS 5.3) rather than a direct path to credential theft, data access, or code execution. No public exploit code was identified at time of analysis, and the practical risk comes from chaining discovered usernames with brute-force, password-spraying, or social-engineering campaigns.
Static-file disclosure in Hono's serveStatic middleware (versions 4.12.4 through 4.13.10) lets remote attackers read files that routing-level access controls were meant to protect, by requesting a malformed percent-encoded path such as '/static/%%36%31dmin/secret.txt' that only normalizes to the protected path during serveStatic's second decode. Affected deployments are those that guard a sub-prefix of the static root with upstream middleware (for example app.use('/static/admin/*', auth) in front of serveStatic('/static/*')); attackers gain read-only access to files already inside the configured static root, with no traversal outside it and no write or availability impact, giving CVSS 3.1 5.3 (AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N) with authentication requirements not applicable per the PR:N vector. No public exploit code or CISA KEV entry was identified at time of analysis; exploitation requires the specific non-default middleware-on-sub-prefix layout, so real-world risk is a moderate, conditional information-disclosure issue rather than a critical one.
FlyCms (through commit edfdf724) fails to enforce object-level authorization in the /ucenter/action/{modelCode}/{id}/counter and /ucenter/action/compose endpoints, letting any registered user tamper with numeric fields - price, stock, points - on content rows belonging to other accounts. The root cause is ActionService.counter(), which calls a checkRowOwner routine that never actually compares the requesting user to the row's owner. Exploitation requires only a self-registered account (authenticated, PR:L) and does not expose data or disrupt availability, so the practical damage depends on whether business-critical numeric fields are in use. No public exploit code or active exploitation was identified at time of analysis.
Authenticated non-owner team members in BoxyHQ Enterprise SaaS Starter Kit through 1.6.0 can perform billing actions reserved for team owners because the Stripe payment API routes skip the team_payments permission check. MEMBER and ADMIN accounts can create portal links and checkout sessions, list products, view invoices, change payment methods, and alter subscriptions on teams that use the built-in Stripe integration. No public exploit code or confirmed active exploitation was identified at time of analysis, and the flaw is limited to provisioned members of the targeted team.
Sensei LMS WordPress plugin versions before 4.26.4 contain a broken access control flaw in the legacy save_new_message handler, allowing any authenticated user, including an unprivileged Subscriber, to create a private message addressed to the author of an arbitrary course, lesson, or quiz post. The handler checks only a generic nonce and does not verify course enrollment or restrict post_id to valid post types, bypassing the enrollment checks in the newer REST send-message controller. Impact is limited to message/spam injection with low integrity impact; no public exploit identified at time of analysis and real-world risk is low-to-moderate.
Cross-site request forgery in Locust (through 2.46.7) lets a remote attacker stop running load tests and wipe recorded statistics in the web UI by luring an operator to a malicious page that issues GET requests to /stop and /stats/reset. No attacker authentication is required, but exploitation depends on operator interaction and on the Locust web UI being reachable from the victim's browser; impact is confined to integrity and availability of test runs and results, with no code execution or data disclosure (CVSS 4.0 score 5.3). No public exploit code was identified at time of analysis, and this is best treated as a low-to-moderate defense-in-depth flaw rather than an urgent threat.
Dolibarr up to and including 23.0.2 uses a weakened Content-Security-Policy on its document-serving endpoints: htdocs/document.php and htdocs/viewimage.php define MAIN_SECURITY_FORCECSP with form-action 'self' instead of the stricter form-action 'none'. This is a defense-in-depth protection-mechanism failure, not a directly exploitable injection flaw; meaningful abuse generally requires a separate content-injection primitive to place a malicious form into content served from those endpoints, which constrains real-world exploitability. The CVSS vector requires authenticated access (PR:L) and no user interaction, and the issue is assessed as moderate-to-low priority; no public exploit was identified at time of analysis, and upgrading to 24.0.0 resolves it.
Local file disclosure in docling-core through 2.102.0 allows remote attackers to make crafted DoclingDocument JSON or YAML embed readable local images as base64 when the document is exported with ImageRefMode.EMBEDDED or rendered to HTML. The bypass supplies plain filesystem paths in ImageRef URIs, evading the allow_image_file_uri gate that in the vulnerable code only guarded file:// URIs. Impact is limited to PIL-parseable images within settings.max_image_decoded_size, though path existence can also be inferred from error behavior; no public exploit was identified at time of analysis.