Use-after-free in the Zephyr RTOS Bluetooth GATT client stack allows an adjacent Bluetooth peer acting as a GATT server to crash a Zephyr device or potentially redirect its execution by replying to a CCC-write with an ATT Error Response. The defect is a callback-ordering bug in gatt_write_ccc_rsp(): the API-documented termination signal notify(NULL) was emitted before subscribe(), so applications that free or recycle bt_gatt_subscribe_params in their termination handler dereference a dangling function pointer. No public exploit code or CISA KEV entry exists; a vendor patch commit is available but a specific tagged release version has not been confirmed.
Unauthenticated information disclosure in ComfyUI (comfy-org/comfyui) before 0.28.0 lets remote attackers read arbitrary image-decodable files and enumerate the host filesystem via the get_model_preview endpoint, which joins an unrestricted {filename:.*} route capture to the model directory with no containment check (CWE-22). Attackers can use literal ../, percent-encoded %2e%2e%2f, absolute paths, or an unbounded path_index to escape the model folder; disclosure is bounded to files Pillow can decode plus a file-existence/path-enumeration oracle. No public exploit identified at time of analysis, and it is not listed in CISA KEV; the issue is fixed in version 0.28.0.
Remote unauthenticated denial of service in 389 Directory Server (389-ds-base, shipped as Red Hat Directory Server 11-13 and within RHEL 6-10) allows an attacker to crash the LDAP daemon by sending a crafted replication extended operation. The flaw is a stack buffer overflow in the RUV parsing path that triggers during payload decoding before any authorization check, so no credentials are needed. No public exploit identified at time of analysis; stack protectors constrain the outcome to a server crash rather than code execution.
Information disclosure in 389 Directory Server (upstream of Red Hat Directory Server 11-13, shipped across RHEL 6-10) lets an unauthenticated remote attacker inject LDAP search filters into the CleanAllRUV replication status-check extended operation. Because the handler evaluates attacker-controlled filters against cn=config using elevated replication-plugin privileges and leaks a boolean match result, attackers can blindly enumerate sensitive configuration metadata such as replication bind DNs and password storage scheme details. No public exploit is identified at time of analysis and it is not listed in CISA KEV, but the network-reachable, no-authentication attack surface makes it a meaningful reconnaissance primitive.
Path traversal arbitrary file read in ComfyUI (comfy-org) before 0.28.0 lets remote unauthenticated attackers probe arbitrary host filesystem paths and exfiltrate image-format files via the /view endpoint. A crafted POST /prompt workflow using LoadImage or sibling loader nodes joins attacker-controlled annotated filenames to a base directory with no containment check, and LoadImage's VALIDATE_INPUTS method causes the engine to skip input-directory validation. Rated CVSS 7.5 (confidentiality-only); no public exploit identified at time of analysis and it is not on CISA KEV.
Denial of service in gnome-remote-desktop as shipped in Red Hat Enterprise Linux 8, 9, and 10 lets an unauthenticated remote attacker exhaust server resources when the daemon runs in system mode with RDP enabled. Because the incoming connection handler bypasses the connection throttler, an attacker can open large numbers of parallel pre-authentication RDP connections, accumulating accepted sockets and pending routing-token operations until timeout and blocking legitimate RDP logins. There is no public exploit identified at time of analysis, and the flaw is specific to Red Hat's build - it does not affect the upstream project.
Pterodactyl Wings before v1.13.0 crashes when its SFTP handler receives a crafted packet with fewer than 4 bytes in the payload during the SSH connection handshake, triggering a Go runtime index-out-of-bounds panic that kills the Wings daemon. The crash is exploitable by any unauthenticated network attacker who can reach the SFTP port, requiring no credentials because the vulnerable slice operation executes prior to authentication. No public exploit code exists at time of analysis, but the patch diff makes the trigger trivially reproducible: any packet with a short payload in the SSH subsystem channel request is sufficient.
Copier CLI and library versions 9.5.0 through 9.16.0 expose a trust-prefix bypass that permits unsafe template features from untrusted repositories to execute on a victim's system after user interaction. The vulnerability arises because Copier's trusted-repository prefix check evaluates the raw percent-encoded URL string, while the underlying HTTP or Git transport subsequently decodes it - meaning a URL containing encoded dot-dot segments (e.g., %2e%2e) or encoded separators (%2f, %5c) can appear to match a trusted prefix while actually resolving to an entirely different, attacker-controlled repository. No active exploitation has been confirmed (not in CISA KEV) and no public exploit code was identified in the provided data; the issue is patched in v9.17.0.
IPv6 packet forwarding in Zephyr RTOS versions 1.8.0 through 4.4.1 never decremented the hop limit on routed unicast packets, directly violating RFC 8200's mandatory loop-termination mechanism and enabling indefinite packet circulation in routing loops. Both forwarding branches of ipv6_route_packet() in subsys/net/ip - the explicit-route path (net_route_packet()) and the on-link cross-interface path (net_route_packet_if()) - were affected, with neither performing a hop-limit check nor decrement before calling net_send_data(). An on-path attacker who can induce or exploit a transient L3 routing loop converts it into a permanent forwarding storm, exhausting CPU and link bandwidth on the Zephyr forwarder and adjacent segments; no public exploit or active exploitation has been identified, and SSVC rates exploitation as none with a non-automatable attack.
Remote code execution via stack-based buffer overflow in Advantech ADAM-3600 EdgeLink's Cookie parsing logic affects all firmware versions prior to 2.8.5.1, exploitable by unauthenticated network attackers when the SafeEnhancement feature is enabled. The vulnerability (CWE-121) allows arbitrary code execution on the edge controller, an OT/IIoT device typically deployed in industrial automation environments. No public exploit has been identified at time of analysis, and EPSS sits at 0.14% (4th percentile), though the industrial attack surface warrants heightened concern due to the operational consequences of device compromise. Notably, the provided CVSS vector (C:H/I:N/A:N) is inconsistent with the described code execution impact - a discrepancy that should be verified with the vendor.
Remote code execution in Advantech ADAM-3600 EdgeLink industrial edge controllers is possible through a stack-based buffer overflow in the device's log message handling functionality, exploitable by unauthenticated network attackers on versions prior to 2.8.5.1. Reported by the Cyber Security Agency of Singapore (CSA), the vulnerability carries a CWE-121 classification - a well-understood root cause class with a history of weaponizable exploits, particularly in embedded and OT/ICS devices. No public exploit has been identified at time of analysis, and EPSS sits at 0.14% (4th percentile), though the network-accessible, unauthenticated nature of the flaw and the industrial deployment context elevate operational risk above what the percentile alone suggests.
Stack-based buffer overflow in Advantech ADAM-3600 EdgeLink's password handling routine enables unauthenticated remote code execution when the SafeEnhancement feature is active. All firmware versions prior to 2.8.5.1 are affected. No public exploit or CISA KEV listing has been identified at time of analysis, but the network-accessible, zero-authentication attack surface on an industrial IoT edge device warrants prompt patching given OT environment exposure.
Quadratic ReDoS in NLTK ReviewsCorpusReader allows an unauthenticated attacker to hang any application that processes untrusted reviews corpora, by supplying a single crafted line of approximately 100,000 bracket-less words. All NLTK versions through 3.9.4 are affected; the calls reviews(), features(), and sents() are all vulnerable entry points. Publicly available exploit code (PoC) is included in the advisory; no confirmed active exploitation in CISA KEV at time of analysis.
Path traversal in NLTK's NKJPCorpusReader component (versions ≤ 3.9.4) allows attackers who control the `fileids` argument to read arbitrary files outside the configured corpus root, fully defeating the `nltk.pathsec` sandbox even when `ENFORCE=True` is active. The root cause is that `add_root()` builds file paths via plain string concatenation - not the hardened `FileSystemPathPointer` - and downstream code calls the Python builtin `open()` directly, bypassing the `validate_path()` check that NLTK's own SECURITY.md explicitly promises will block unauthorized access. A detailed public PoC is confirmed against nltk 3.9.4; no active exploitation is listed in CISA KEV at time of analysis.
Path traversal in NLTK's FramenetCorpusReader (versions ≤3.9.4) allows a caller-supplied frame name containing ../ sequences to escape the corpus root and read arbitrary XML files accessible to the process, silently bypassing the nltk.pathsec sandbox even when ENFORCE=True is explicitly enabled. Any web service or pipeline that routes user-controlled input into FramenetCorpusReader.frame() is directly exposed, with full parsed content of frame-shaped XML files returned to the caller and a file-existence oracle for all other XML paths. A working proof-of-concept is publicly available in the GitHub security advisory; the vulnerability is not listed in CISA KEV at time of analysis.
Command injection in Cudy TR1200 (firmware v2.4.15) and TR3000 (firmware v2.4.21) routers via the system.setclock interface enables an authenticated administrator to execute arbitrary OS commands with root privileges. The vulnerability stems from unsanitized input passed directly to a system command, yielding full device compromise - confidentiality, integrity, and availability all rated high. No public exploit or active exploitation (CISA KEV) has been identified; EPSS of 1.21% at the 65th percentile reflects moderate relative risk among disclosed CVEs but low absolute exploitation probability.
Privilege escalation in FeehiCMS v.2.1.1 exploits insufficient session expiration (CWE-613) across the session management module, authentication logic, and logout handler, allowing a network-based attacker to reuse captured session tokens even after legitimate user logout to gain elevated access. The CVSS 3.1 score of 7.4 (High) reflects high confidentiality and integrity impact under high attack complexity, with no authentication required on the attacker's side. A public proof-of-concept write-up exists on GitHub; however, EPSS sits at 0.20% (10th percentile) and no CISA KEV listing exists, suggesting limited observed exploitation despite available exploit documentation.
Authenticated command execution affects several Hikvision wireless access points (DS-3WAP521-SI, DS-3WAP522-SI, DS-3WAP621E-SI, DS-3WAP622E-SI, DS-3WAP623E-SI, DS-3WAP622G-SI), where insufficient input validation lets an attacker holding valid device credentials send crafted packets carrying malicious commands to achieve arbitrary command execution on the device. The CVSS 3.1 vector (7.2, AV:N/AC:L/PR:H/UI:N) indicates a network-reachable but high-privilege (authenticated) attack with full confidentiality, integrity, and availability impact. There is no public exploit identified at time of analysis and the issue is not listed in CISA KEV; the vendor has published a security advisory.
Session hijacking in Coturn TURN/STUN server prior to 4.15.0 allows an authenticated attacker who obtains a victim's MOBILITY-TICKET to seize the victim's TURN allocation, intercepting and injecting relayed traffic while exhausting the victim's quota. The flaw resides in the --mobility (MICE, RFC 8016) session-resume code path in handle_turn_refresh: the server locates the victim allocation via an attacker-controlled mobile ID and then validates the REFRESH against the attacker's own credentials rather than the allocation owner's, because the hmackey_set flag set during the attacker's own prior authentication is never cleared for long-term-credential sessions and silently suppresses the credential-copy step. No public exploit code or CISA KEV listing is identified at time of analysis; CVSS AC:H and PR:L reflect that exploitation requires both a legitimate authenticated session and possession of the victim's MOBILITY-TICKET.
Default-credential unauthorized access affects the Advantech ECU-1251D industrial communication gateway, which ships with a root SSH account protected by no password and is documented as such in the device manual. An attacker on the same network segment can log in as root over SSH without any credential, and because configuration provides no prompt or guidance to change these defaults, exploitable devices are likely to persist in the field. No public exploit identified at time of analysis, but the vulnerability requires no exploitation skill - knowledge of the documented default is sufficient.