CVE Feed

Ivanti Connect Secure 22.8R2.4 mishandles the manner in which the dsinvoked broker trusts DSCache configuration data, leading to privilege escalation to root, as exploited in the wild in October 2026.

The EnergizeLab Eilik NestJS auth service accepts a user-controlled user ID in account update requests (mass assignment). A registered user can set another account's identifier and take over that account.

EnergizeLab station-middleware (NestJS) Docker images 1.6.5 through 1.8.2 do not check user roles on administrative APIs. Any authenticated low-privilege user can retrieve AI provider keys and change configuration and object-store contents.

The EnergizeLab Eilik and EilikO FastAPI AI inference and profiling services expose AI model administration functions without authentication. An unauthenticated remote attacker can change AI model configuration and switch the model in use at runtime.

EnergizeLab station-middleware (NestJS) Docker images 1.6.5 through 1.8.2 allow any registered, low-privilege account to read arbitrary files from the server by supplying an attacker-controlled file path. This exposes middleware files and server secrets.

EnergizeLab AI Station firmware, EilikO Linux robot firmware v2.1_20260617182218_release, and the EnergizeLab Windows desktop app 3.1.2 download firmware updates over plain HTTP and do not verify their integrity or authenticity. An attacker able to intercept or modify the traffic can substitute a malicious update.

The EnergizeLab FirmwareUpdater (AArch64 ELF, distributed in the CDN bin.zip) uses a single hard-coded AES key shared by all installations. An attacker who recovers the key from the updater can decrypt and forge firmware packages for any device that trusts it.

The EnergizeLab Eilik NestJS auth service issues device tokens for arbitrary robot serial numbers without authentication. An unauthenticated remote attacker can obtain a valid device token for any serial number and act as that device against the cloud services.

The EnergizeLab AI Station (RK3562) firmware image ships with default root credentials and SSH enabled. An attacker with network access to the device can log in as root and take full control of it.

The EnergizeLab Eilik and EilikO AI skill upload endpoint, served by the FastAPI AI inference and profiling services, does not require authentication and does not constrain uploaded file paths. An unauthenticated remote attacker can use path traversal to write arbitrary files, and thereby tamper with AI skill files used by Eilik and EilikO robots.

This CVE ID has been rejected or withdrawn by its CVE Numbering Authority.

The vulnerability exists within internal components of BeyondTrust’s Remote Support product. The components did not adequately validate the origin of the messages it received. Under the right conditions, this could allow an attacker to perform file operations on the appliance.

ZoneMinder is a free, open source closed-circuit television software application. In versions 1.34.0 through 1.36.38, the web interface mishandles failed token authentication in web/includes/auth.php. validateToken() returns false when a token is malformed or has an invalid signature, but the caller assigns that value to the global $user variable; because web/index.php tests isset($user), boolean false is treated as a present user and the normal login redirect is skipped. This can dispatch views that rely on the outer authentication gate, including the zones view, which accepts an attacker-selected monitor ID and discloses monitor and zone metadata without an independent permission check. Permission helpers receive the false user value and deny privileged operations, so unauthorized state changes have not been established. This vulnerability is fixed in version 1.36.63, where conversion to ZM\User objects causes invalid token state to fail before view dispatch. Later defense-in-depth changes explicitly leave $user unset after failed token validation and reject non-ZM\User values at the web authentication gate.

temporal-proxy 0.2.0 through 0.8.0 does not bind encrypted payloads to the namespace they were encrypted for. When payload encryption is enabled, the proxy chooses the decryption key by an identifier stored in the payload rather than by the namespace of the request, and nothing in the ciphertext records that namespace. A caller whom the proxy authorizes only for namespace B, and who has obtained an encrypted payload from namespace A, can therefore have the proxy decrypt it. In 0.8.0, the caller can submit the payload to the codec server's /decode endpoint. In every affected version, a caller who can also write to namespace B on the Temporal server directly can place the payload there and read it back through the proxy. This holds whether namespaces share the default key or each has its own, because a single proxy holds every key it is configured with. The issue does not give the attacker encrypted payloads they could not already obtain, and does not allow encrypted payloads to be modified.

An authorization flaw in Idira Conjur version 13.9.0 allows an authenticated user with limited privileges to perform unauthorized actions on a policy branch. Under specific circumstances, an attacker could potentially leverage this to retrieve unauthorized secrets, alter administrative access controls on the branch or cause a denial of service (DoS). Idira Security Bulletin CA26-38.

@casl/ability 6.7.5 before 7.0.2 contains a prototype pollution vulnerability in rulesToFields() that allows attackers to modify inherited builtins via dotted condition keys. Attackers controlling rule conditions can supply keys like 'toString.call' to alter shared builtin functions process-wide, causing denial of service for the Node.js process.

Web3.py 7.15.0 through 7.16.0 and 8.0.0b2 through 8.0.0 contains a server-side request forgery vulnerability because _check_ip_blocked() does not unwrap IPv4-mapped IPv6 addresses in CCIP Read URL validation. Malicious contracts can return OffchainLookup URLs with IPv4-mapped IPv6 hosts or AAAA records, making clients with CCIP Read enabled send requests to loopback, private, or link-local services.

kedro-datasets from 9.3.0 through 9.6.0 contains a path traversal vulnerability in MatplotlibDataset.save() that allows attackers to write files outside the dataset directory by supplying absolute plot names. Because validate_sub_path() strips leading slashes while pathlib's join discards the base path, attackers can create or overwrite files anywhere the process can write.

La Suite Docs through 6.0.1 contains an authorization bypass vulnerability that allows authenticated users to write comments into threads of documents they cannot access. Attackers who know a thread UUID can route a POST comment request through their own document's resource_id, inserting comments into another document's thread without access.

LibreDesk before 2.9.0-rc.8 contains a stored cross-site scripting vulnerability that allows authenticated agents to execute script by uploading SVGs with malformed Content-Type parameters. Attackers can upload an SVG typed 'image/svg+xml; x' via /api/v1/media so GET /uploads/{uuid} serves it inline without sandbox, running script in any viewing agent's or administrator's session.

GNU LibreDWG through 0.14.8612 contains a global out-of-bounds read vulnerability in the AcDs decoder acds_private() that allows attackers to read beyond a static table via an unchecked schema property type index. Attackers can supply a crafted R2004 or R2010+ DWG file with a large 'schdat' property type value to crash dwgread, dwg2dxf, or any application using LibreDWG.

libmikmod through 3.3.15 contains a heap out-of-bounds read vulnerability in the IMF_Load() loader that allows attackers to read beyond allocated memory by supplying a crafted IMF module. Attackers can enable the linear-period header flag and map notes to out-of-range samples, causing noteindex indexing far past the heap allocation and crashing applications that load the file.

libmikmod through 3.3.15 contains a global out-of-bounds read vulnerability in the MultiTracker loader MTM_Load() because the module channel count is never clamped. Attackers can supply a crafted .mtm file with a channel count above 32 to read past the static pat array, corrupting track indices and crashing the loading application.

UADE through 3.05 contains an out-of-bounds read vulnerability in uade_filemagic() in amifilemagic.c caused by unsigned bufsize subtraction wrapping on short inputs. Attackers can supply a crafted short module that passes format probe bounds checks and reads far past the buffer, crashing applications that probe or play it.

WildMIDI through 0.5.0 contains a heap out-of-bounds read vulnerability in _WM_ParseNewHmi() in src/f_hmi.c that reads past buffers when handling 0xfe sub-events. Attackers can supply a crafted HMI file placing a 0xfe 0x10 sub-event near the buffer end to read up to four bytes past the allocation, crashing the decoding application.

WildMIDI through 0.5.0 contains a heap out-of-bounds read vulnerability in the HMP parser _WM_ParseNewHmp() that allows attackers to read beyond buffers via underflowed chunk lengths. Attackers can supply a crafted .hmp file with a chunk length of 0-12 that wraps to roughly 4 GB, crashing applications or exposing heap bytes through WildMidi_GetInfo().

libADLMIDI 1.6.2 through 1.6.2.1 contains an off-by-one stack out-of-bounds write vulnerability in BW_MidiSequencer that allows attackers to corrupt stack memory via crafted MIDI files. Attackers can supply a MIDI file with a note-on using status 0x9F and note 0x7F through adl_openFile() or adl_openData() to crash the hosting media player or game engine.

LightLLM through 1.2.0 contains a denial of service vulnerability that allows unauthenticated attackers to crash the inference worker by supplying the internal pd_kv_trans_params field to the /generate endpoint. Attackers can send a single request whose parameters cause pickle.loads to fail in InferSamplingParams, killing the inference thread while /health stays up and all later requests hang until restart.

libewf through 20260924 contains a heap out-of-bounds read vulnerability in libewf_lef_extended_attribute_read_data() because UTF-16 name and value sizes are validated before being doubled. Attackers can supply a crafted .L01/.Lx01 file with oversized 'ea' metadata counts to crash the application or leak uninitialized heap memory through extended attributes.

EarthScope libmseed through 3.5.4 contains a heap out-of-bounds read vulnerability in ms3_detect() in parseutils.c that reads past truncated miniSEED 2 record buffers. Attackers can supply a 40 to 47 byte buffer passing MS2_ISVALIDHEADER to msr3_parse() or ms3_detect(), over-reading up to 8 bytes and potentially crashing the process.

SoX through 14.4.2 contains a heap out-of-bounds write vulnerability in lsx_ms_adpcm_block_expand_i() in src/adpcm.c, triggered by an odd samplesPerBlock value in the WAV fmt chunk. Attackers can supply a crafted MS ADPCM WAV file to sox or any libsox consumer to write a partially controlled 16-bit sample past the heap buffer allocated by wav_ms_adpcm_fmt().

libTiMidity through 0.2.8 contains an out-of-bounds access vulnerability in init_soundfont() in src/sndfont.c that allows attackers to corrupt memory by supplying unvalidated SF2 preset header bank and preset values. Attackers can craft a .sf2 file whose bank or preset indexes past the 128-entry tonebank or drumset arrays, causing heap memory reads, writes, and application crashes.

libTiMidity through 0.2.8 contains a heap out-of-bounds read vulnerability in the SoundFont loader in src/sndfont.c that uses unvalidated sampleId and instrument generator values as table indices. Attackers can supply a crafted .sf2 file loaded through mid_set_soundfont() to read heap memory out of bounds and crash the application.

libTiMidity through 0.2.8 contains a stack out-of-bounds write vulnerability in the SoundFont loader that allows attackers to corrupt stack memory via unchecked generator operators in parse_gen. Attackers can supply a crafted .sf2 file with out-of-range pgen/igen operator values to write chosen 16-bit values near the stack Layer object, causing crashes and stack memory corruption.

Game Music Emu (libgme) through 0.6.5 contains a global out-of-bounds read vulnerability in Classic_Emu::mute_voices_ that allows attackers to crash applications by supplying NSF files whose voice count exceeds the 16-entry voice-type table. Attackers can craft an NSF header with chip_flags 0x3F, opened through gme_open_data or gme_open_file, to read past the static array during loading and cause denial of service.

Game Music Emu (libgme) through 0.6.5 contains a heap out-of-bounds read vulnerability in the AY loader get_data function caused by an integer underflow in the track-table bounds check. Attackers can supply a crafted ZXAYEMUL file with an oversized track table to read past the file buffer via gme_track_info or gme_start_track, crashing libgme consumers.

OpenJPH through 0.32.0 contains a heap out-of-bounds write vulnerability in the HTJ2K cleanup-pass decoder that allows attackers to corrupt memory via one-row codeblocks. Attackers can supply a crafted .j2c or .jph codestream with small precinct sizes, causing the decoder to write a second row past the codeblock buffer, resulting in memory corruption and crashes.

ValleyBell libvgm through commit a6eed945 contains a heap out-of-bounds read vulnerability in the PCM data-block decompressor caused by an unbounded compression-table index. Attackers can supply a crafted .vgm file with a small type 0x7F table and wider compressed values, causing DecompressDataBlk to over-read heap memory and crash consumers.

FreeImage (danoli3 fork) through 3.19.23 contains a heap out-of-bounds read vulnerability in Unpack32Bits() in PluginPICT.cpp because rowBytes is never checked against the file-controlled PixMap width. Attackers can supply a crafted PICT file with a truncated or undersized rowBytes to over-read up to 64 KiB of heap, leaking memory into decoded pixels or crashing the process.

libsidplayfp through 3.1.1 contains a heap out-of-bounds read vulnerability that allows attackers to crash applications by supplying a PSID/RSID file with an unchecked data offset. Attackers can set the header data offset beyond the file size, wrapping the length calculation so resolveAddrs() reads up to 64 KiB past the heap buffer.

Libyal libesedb through 20260704 contains a heap out-of-bounds read vulnerability in libesedb_record_get_multi_value() that allows attackers to read past buffers via unbounded multi-value offset tables. Attackers can supply a crafted ESE/EDB database processed by esedbexport to crash the application or expose up to about 32 KiB of adjacent heap memory.

ytnef through 2.1.2 contains a heap out-of-bounds read vulnerability in MAPIPrint() that allows attackers to read past heap allocations by supplying compressed RTF properties. Attackers can deliver a crafted winmail.dat whose decompressed RTF lacks a NUL terminator, causing printf to over-read, crashing ytnefprint and leaking adjacent heap bytes.

ytnef through 2.1.2 contains a heap out-of-bounds read vulnerability in DecompressRTF() in lib/ytnef.c that never validates uncompressedSize for uncompressed RTF streams. Attackers can supply a crafted winmail.dat TNEF attachment with a malicious PT_BINARY MAPI property to crash ytnefprint or libytnef callers and leak adjacent heap memory.

libansilove 1.3.0 through 1.4.2 contains a NULL pointer dereference vulnerability in ansilove_xbin() that allows attackers to crash applications by supplying a crafted XBIN file. Attackers can set the 512-character flag without the custom-font flag, causing drawchar() to index a NULL high_font_data_xbin pointer and terminate the process.

Ansilove libansilove through 1.4.2 contains an out-of-bounds read vulnerability in ansilove_xbin() in src/loaders/xbin.c that lacks bounds checks for compressed run-length records. Attackers can supply a crafted compressed XBIN file ending in a control byte such as 0x7F to read past the input buffer, crashing applications that render untrusted XBIN art.

zziplib through 0.13.80 contains an out-of-bounds read vulnerability in __zzip_parse_root_directory() in zzip/zip.c that allows attackers to read past the mapped central directory in mmap builds due to an ineffective bounds check. Attackers can supply a crafted ZIP whose entry name ends at the central directory end, triggering 28-byte ZIP64 extra field over-reads via zzip_dir_open() that crash the process.

Zziplib through 0.13.80 contains an out-of-bounds read vulnerability in zzip/mmapped.c that allows attackers to over-read heap memory via a crafted ZIP central directory. Attackers can craft an end-of-central-directory record placing the first entry near the buffer end, causing zzip_disk_entry_strdup_name to read past the buffer and crash the application.

libregf through 20260526 contains a heap out-of-bounds read in libregf_value_item_read_value_data() because it bounds value data size without subtracting the 4-byte old-format cell offset. Attackers can supply a crafted version 1.0/1.1 REGF hive with a value data cell ending its hive bin to leak adjacent heap bytes or crash parsers like regfexport and plaso.

NetSurf libnsbmp through 0.1.7 contains an out-of-bounds read vulnerability in bmp_decode_mask() that allows attackers to read one byte past the image buffer by supplying crafted ICO files. Attackers can provide an ICO whose inner bitmap width is not a multiple of 8, causing a heap over-read that can crash applications calling bmp_decode().

NetSurf libnsbmp through 0.1.7 contains an out-of-bounds read vulnerability in bmp_decode_rle8() and bmp_decode_rle4() because absolute-run bound checks omit the word-alignment pad byte. Attackers can supply a crafted RLE8 or RLE4 BMP ending on an odd-length absolute run to cause a one-byte heap over-read, crashing applications decoding untrusted images.