CVE-2026-119342 is an improper input validation vulnerability in Stirling-PDF versions 2.13.1 up to but not including 2.13.2. The flaw exists in the POST /api/v1/convert/svg/pdf endpoint, which renders attacker-supplied SVG content using Apache Batik. A scheme-less root-relative image reference bypasses the SVG sanitizer, allowing local file disclosure of image-decodable files on the server. This can occur without authentication on login-disabled deployments or with non-admin access when login is enabled. The vulnerability discloses PNG, JPG, GIF, or SVG content but does not disclose plain text, binary files, directories, or external HTTP references. The issue is fixed in version 2.13.2.
CVE-2026-62853 is a critical OS command injection vulnerability in the mqttsnet ThingLinks enterprise IoT platform prior to version 1.4.0. The vulnerability exists in the /groovyScriptTest/registerScript endpoint, which accepts user-supplied Groovy scripts and compiles them without sandboxing. An authenticated attacker can exploit this to execute arbitrary code on the server. This flaw affects the rule engine's Groovy script test and register flow and is fixed in version 1.4.0.
CVE-2026-117645 is a cross-site scripting (XSS) vulnerability in the phoenix_html library of the phoenixframework. It arises from improper neutralization of input during web page generation, specifically in the handling of attribute names in HTML rendering functions. The vulnerability affects phoenix_html versions from 1.0.0 up to but not including 3.3.5, and from 4.0.0 up to but not including 4.4.0. The issue allows an attacker controlling attribute names to inject malicious attributes and scripts into rendered HTML. This occurs because attribute names are not properly escaped, enabling injection via constructs like HEEx <tag {@attrs}>, content_tag/3, or direct calls. Fixed versions are 3.3.5 and 4.4.0.
CVE-2026-119341 is a vulnerability in the open source OCR engine Tesseract prior to version 5.5.4. It involves improper handling of pointers when reading ADAPT_CLASS_STRUCT data from pre-adapted-template files, which can lead to memory corruption and denial of service. The issue is fixed in version 5.5.4.
The pypi-publish GitHub Action versions before 1.14.3 included the emSign root CA certificate via the certifi bundle, which has since been removed. This change affects the TLS trust store used by Sigstore for Trusted Publishing and Digital Attestations. The vulnerability is tracked due to the revocation of this CA in the chain that Sigstore relies on. A fix is available in version 1.14.3.
Metalama is a Roslyn-based open-source patterns and architecture toolkit for C#. Prior to 2025.1.18, 2026.0.25 and 2026.1.19, Metalama.Framework and Metalama.Tool load and execute compile-time assemblies, auxiliary executable tools and embedded helper assemblies from predictable temporary directories. On `Linux` and `macOS`, the default `Path.GetTempPath()` location is world-writable, and existing artifacts are trusted based on sentinel or `.completed` marker files rather than integrity verification. A local unprivileged user on a shared machine or CI/build agent can pre-create the deterministic directory structure and plant a malicious DLL with the matching sentinel. When a higher-privileged developer or build account subsequently builds, Metalama skips regeneration and executes the planted artifact in the compiler, IDE or MSBuild process as the building user. This permits cross-user privilege escalation, reading source and secrets, supply-chain tampering with build output, or disrupting builds. Default Windows per-user temporary directories are protected by ACLs and are not affected. Deployer-selected insecure cache locations are outside the vulnerability's scope. This issue is fixed in versions 2025.1.18, 2026.0.25 and 2026.1.19.
CVE-2026-119340 is an integer overflow or wraparound vulnerability in the Tesseract OCR engine prior to version 5.5.4. The issue arises from the GENERIC_2D_ARRAY::DeSerializeSize function reading matrix dimensions as signed 32-bit integers but only validating them against UINT16_MAX. This allows negative or excessively large dimension values to cause inconsistent memory allocation states, leading to out-of-bounds accesses when loading crafted traineddata files. The vulnerability can result in denial of service. The issue is fixed in Tesseract version 5.5.4.
CVE-2026-117666 is a critical vulnerability in the deadtrickster ssl_verify_fun library affecting versions from 1.0.1 up to but not including 1.1.8. The flaw involves improper validation of TLS certificates with hostnames containing partial-label wildcards, allowing an attacker in a man-in-the-middle position to impersonate a TLS peer by exploiting the wildcard matching logic. This occurs because the wildcard matching function incorrectly allows wildcards to consume dots, violating RFC 6125's restriction that wildcards only apply to the left-most label. The issue affects clients that disable Server Name Indication (SNI), servers verifying client certificate hostnames with this library, and direct calls to verify_cert_hostname/2. Erlang/OTP 20.0 and later mitigate this by checking hostnames before user verification functions if SNI is enabled. The vulnerability is fixed in ssl_verify_fun version 1.1.8.
Metalama, a Roslyn-based toolkit integrated with Visual Studio, contains a deserialization vulnerability in its Windows Visual Studio integration from versions 2023.0.101-rc through 2026.1.18. The vulnerability arises from unsafe deserialization of local named-pipe messages without restricting CLR types, allowing a same-user local process to execute arbitrary code within the Visual Studio process. The issue is fixed in Metalama versions 2025.1.18, 2026.0.25, and 2026.1.19, with the maintained Metalama + PostSharp extension fixed in 2026.1.5. Retired extensions require removal or migration.
CVE-2026-62300 is a resource exhaustion vulnerability in filecoin-pin up to version 1.1.0. The service does not enforce rate limits or quotas on pinning requests, allowing authenticated or anonymous users (if configured) to submit large DAGs that exhaust disk, network, CPU, memory, and libp2p resources. This can degrade service or cause process failure, leaving partial CAR files that require recovery. Default localhost binding and token requirements reduce exposure. Version 1.1.1 documents mitigations but does not implement resource controls.
Tesseract is an open source OCR engine. Prior to 5.5.4, WeightMatrix::DeSerialize trusts a mode byte from a traineddata LSTM network and calls WeightMatrix::DeSerializeOld when kDoubleFlag is clear. In default FAST_FLOAT builds, DeSerializeOld invokes ASSERT_HOST instead of returning an error, so loading a crafted model through TessBaseAPI::Init terminates the process and causes denial of service. This issue is fixed in version 5.5.4.
CVE-2026-58412 is an authorization vulnerability in filecoin-project's filecoin-pin prior to version 1.1.0. The flaw arises from improper fork detection in the GitHub Action upload workflow, allowing an external contributor to cause attacker-controlled artifacts to be uploaded using the repository owner's Filecoin wallet. This leads to unauthorized storage costs, dataset pollution, and consumption of wallet funds without requiring disclosure of the wallet private key. The issue is fixed in version 1.1.0.
CVE-2026-119338 is a medium severity vulnerability in the Tesseract OCR engine versions 5.4.0 through 5.5.3. The flaw involves an out-of-bounds read and use-after-free in the AppendLinePolygon function, which mishandles right-to-left polygon accumulators on certain language pages like Hebrew or Arabic. This leads to double free and corrupted coordinate data. The issue is resolved in version 5.5.4.
CVE-2026-119337 is a missing authorization vulnerability in the open-source RAGFlow engine versions 0.25.2 through 0.27.2. It allows an authenticated user to access another user's private agent by exploiting a flaw in the POST /api/v1/agents/chat/completions endpoint before proper access checks occur. This can lead to unauthorized reading of private agent outputs and potential side effects from configured tools. The issue is fixed in version 0.27.3.
CVE-2026-55098 is a vulnerability in Basekick-Labs' Arc time-series database prior to version 26.06.2. It allows unauthorized actors to bypass cross-database authorization controls via certain unblocked DuckDB table-valued I/O functions. This enables reading and enumeration of other tenants' time-series data and schema information within the allowed storage root. The issue was fixed in version 26.06.2.
CVE-2026-62333 is a high-severity vulnerability in the openvm-org OpenVM zkVM framework affecting versions prior to 1.7.0 in the 1.x line and certain prereleases of 2.0 and 2.1. The flaw involves improper input validation in the MemoryMerkleAir component, allowing a malicious prover to fabricate a subtree below a memory leaf and alter the committed memory root without corresponding memory operations. This compromises the soundness of OpenVM proofs using persistent memory. The issue is fixed in OpenVM versions 1.7.0, 2.0.0-rc.3, and 2.1.0-preview.
OpenVM is a performant and modular zkVM framework built for customization and extensibility. Before 1.7.0 in the 1.x line and in affected 2.0 and 2.1 prereleases, `OpenVmHalo2Verifier.verify` constructs raw proof calldata from user-controlled `proofData`, performs a `staticcall` to `address(this)`, and treats a successful call as proof acceptance. If the first four proof bytes match another external selector exposed by the wrapper, dispatch can invoke a successful non-verifier function instead of the generated Halo2 fallback verifier. Applications using the generated Solidity wrapper can consequently accept arbitrary app commitments and public values without checking the proof suffix, accumulator, app commitments or public values. This issue is fixed in versions 1.7.0, 2.0.0-rc.3 and 2.1.0-preview.
Pi-hole is a DNS sinkhole that protects devices from unwanted content without installing any client-side software. From 6.4.2 until 6.5, the root-run advanced/Scripts/updatecheck.sh cron job applies truncate -s 0 and chmod 644 to /etc/pihole/versions while following symbolic links in a directory owned by the pihole user. An attacker executing as pihole can replace that path with a symlink so the scheduled root operations truncate or create an arbitrary root-owned file, or race the operations so chmod makes a sensitive target world-readable without truncating it. This issue is fixed in version 6.5.
CVE-2026-119335 is an OS command injection vulnerability in pi-hole versions prior to 6.5. The vulnerability arises from improper validation of the pihole-owned PID file, allowing an attacker with pihole user privileges to craft a malicious PID file that leads to command execution as root. This flaw bypasses previous remediation measures and is fixed in pi-hole version 6.5.
CVE-2026-62335 is an improper input validation vulnerability in openvm-org's OpenVM zkVM framework. Versions before 1.7.0 in the 1.x line, and certain prereleases of 2.0 and 2.1, allow the verify function to accept non-canonical 256-bit values for app executable and VM commitments without verifying they are less than the BN254 scalar field modulus. This can cause ambiguity between the byte-level inputs and the field elements used by the verifier. The issue is fixed in versions 1.7.0, 2.0.0-rc.3, and 2.1.0-preview.