A vulnerability in the Audio, Web, and Video Conferencing (AWV) component of Mitel MiCollab through 10.3 (10.3.0.18) could allow an unauthenticated attacker to conduct a blind command injection attack due to insufficient parameter sanitization. A successful exploit could allow an attacker to execute arbitrary commands within the context of the system. This is MTLVULN-1864.
CVE-2026-133311 is a medium severity vulnerability in the Audio, Web, and Video Conferencing (AWV) component of Mitel MiCollab through version 10.3.0.18. It allows an unauthenticated attacker who knows a specific system-generated random identifier or pre-staged content to bypass authentication controls and gain unauthorized access to sensitive information within the affected functionality. The vulnerability does not allow access to arbitrary files outside this scope.
A remote code execution (RCE) vulnerability exists in jane-php/open-api-3 before version 7.12.0. The issue arises because PHPDoc comment text is emitted raw without escaping, allowing an attacker to inject a closing comment sequence '*/' in a query parameter name. This prematurely closes the PHPDoc block and injects attacker-controlled PHP code into the generated Client class, which executes upon instantiation. This vulnerability has a CVSS 4.0 score of 9.3 (critical).
CVE-2026-133303 is a critical XML injection vulnerability in the Audio, Web, and Video Conferencing (AWV) component of Mitel MiCollab through version 10.3.0.18. It allows an unauthenticated attacker to leverage pre-staged malicious content due to insufficient parameter validation, potentially enabling arbitrary command execution within the system context.
A remote code execution vulnerability exists in jane-php/open-api-3 before version 7.12.0. The issue arises because header parameter names containing the sequence '*/' break out of PHPDoc comment blocks in generated client code, allowing attacker-controlled PHP code injection. This injected code executes when the generated client class is instantiated, leading to critical remote code execution.
A remote code execution (RCE) vulnerability exists in jane-php/open-api-3 before version 7.12.0. The issue arises because parameter descriptions in generated PHPDoc comments are emitted without escaping the sequence '*/', allowing an attacker to prematurely close the docblock and inject arbitrary PHP code into the generated Client class. This injected code executes when the client is instantiated, enabling attacker-controlled code execution.
A vulnerability in the Audio, Web, and Video Conferencing (AWV) component of Mitel MiCollab through 10.3 (10.3.0.18) could allow an unauthenticated attacker to conduct an arbitrary file upload attack due to insufficient file type validation. A successful exploit could allow an attacker to upload arbitrary files to the target system. This is MTLVULN-1861.
Mareforma version 0.3.3 has a critical integrity vulnerability in its REPLICATED convergence logic, allowing unsigned claims to be promoted to REPLICATED status due to missing signature enforcement. The peer-selection SQL query does not verify cryptographic signatures, and with default settings, unsigned claims automatically meet convergence preconditions. This enables attackers without valid keys to inject forged claims that appear as multi-party endorsements in trust queries.
CVE-2026-71534 is a vulnerability in mareforma versions >=0.3.0 and <0.3.8 where the transparency_logged flag is incorrectly derived from an unsigned Rekor metadata block. This allows an attacker who can edit claims.toml to forge the Rekor-witnessed state without actual logging. Additionally, honest configurations without Rekor suffer silent trust degradation after restore operations, preventing convergence to the REPLICATED state. The issue is fixed in version 0.3.8.
A vulnerability in Mareforma versions 0.3.0 up to but not including 0.3.8 allows invalidated claims to be promoted to the REPLICATED status due to missing checks on the t_invalid flag during claim convergence. This flaw permits a refuted claim to re-enter the trust ladder, potentially promoting an honest claim incorrectly. The issue arises from three gaps in the code where t_invalid is not checked during new claim short-circuiting, candidate-peer selection, and promotion update, creating a time-of-check to time-of-use window. A fix is available in version 0.3.8 that properly incorporates t_invalid checks and closes the TOCTOU window.
The mareforma package's verify_bundle function claims to verify each per-claim signature within a DSSE bundle but does not actually perform cryptographic verification of individual claim signatures. Instead, it only checks internal digest consistency, allowing a malicious or compromised exporter key holder to produce bundles with missing, invalid, or forged claim signatures that still pass verification. This undermines the trust boundary the export mechanism is intended to provide. A fix is available in version 0.3.8.
The Mareforma package contains a vulnerability in the search_claims function where it skips the verify-on-read step, allowing forged or tampered rows with high trust levels to be served without verification. This results in search_claims returning forged ESTABLISHED rows that query_claims correctly excludes, violating the documented tamper guarantee. The flaw arises because search_claims applies only a preliminary filter and does not re-verify signatures on read, unlike query_claims. A fix is available in version 0.3.8 that aligns search_claims behavior with query_claims by adding verification and adjusting batching logic.
An authenticated MQTT client in NATS Server versions v2.14.2, v2.12.7 and below could bypass subscribe permissions to access internal MQTT delivery subjects. This exposure reveals MQTT QoS 2 protocol metadata for sessions within the account, though message payloads remain protected. The vulnerability arises from incomplete restrictions on internal MQTT subjects. Fixed versions are 2.12.12 and 2.14.3. No workaround exists for deployments requiring MQTT; upgrading is necessary.
PHP released security patches addressing seven critical remote code execution (RCE) vulnerabilities affecting multiple PHP versions. These flaws include use-after-free bugs, session deserialization issues, and request smuggling in PHP-FPM, which could allow attackers to execute arbitrary code with the privileges of the PHP worker or cause worker crashes. The vulnerabilities impact PHP versions 8.3 through 8.6, with PHP 8.2 not affected by some issues. PHP 8.6.0 and 8.5.13 include fixes for all critical bugs, and patches are also available for PHP 8.4.28, 8.3.37, and 8.2.36. Exploitation requires specific conditions such as control over session data or request bodies, and some RCE scenarios have not yet been demonstrated. No known exploits in the wild have been reported at this time.
HTSlib is a library for reading and writing bioinformatics file formats. Prior to 1.21.2, 1.22.3, 1.23.2, and 1.24, the CRAM BETA decoder functions cram_beta_decode_long, cram_beta_decode_int, and cram_beta_decode_char, and the experimental CRAM v4 XPACK functions cram_xpack_decode_long and cram_xpack_decode_int, have bounds-checking flaws caused by integer overflow after a multiplication. The resulting incorrect bounds check can allow decoding to read beyond the supplied input buffer, potentially leaking process-state information or crashing the process. This issue is fixed in versions 1.21.2, 1.22.3, 1.23.2, and 1.24.
A divide-by-zero vulnerability exists in littlefs versions 2.8.0 through 2.11.3 in the lfs_mount_() function. This flaw occurs when mounting crafted filesystem images with a block_count of zero, causing a crash due to a SIGFPE signal. The vulnerability can be triggered by supplying an image missing a superblock entry or with block_count set to zero, leading to denial of service.
Quasar before 2.35.3 contains a DOM clobbering vulnerability that allows attackers to break native wrapper detection by injecting HTML elements named Capacitor, cordova, or StatusBar. Attackers can place markup such as a div with id Capacitor into server-rendered or pre-rendered HTML so app.use(Quasar) throws and the application never mounts, or openURL() and AddressbarColor.set() throw.
CVE-2026-133238 is a relative path traversal vulnerability in SABnzbd versions up to 5.1.3. It allows attackers to write files outside the intended Complete folder by supplying crafted NZB files with specially constructed name values that manipulate directory paths. This can lead to files being placed in arbitrary directories writable by SABnzbd.
modified eCommerce Shopsoftware versions up to 3.3.0 contain a local file inclusion vulnerability. This flaw allows an attacker with administrator privileges to include arbitrary server files by exploiting improper control of filenames in the content manager. Specifically, by storing '../' sequences in file names, unauthenticated requests to shop_content.php can cause the server to include and execute PHP code from uploaded content files or images as the web server user.
modified eCommerce Shopsoftware through 3.3.0 contains a reflected cross-site scripting vulnerability in admin/support.php that allows unauthenticated attackers to inject script via the module parameter. Attackers can trick a logged-in administrator into opening a crafted link to run script in the admin panel, reading CSRF tokens and changing shop settings.