PYSEC-2026-2656 Arbitrary file write via tar traversal in mlflow

A vulnerability in MLflow's pyfunc extraction process allows for arbitrary file writes due to improper handling of tar archive entries. Specifically, the use of tarfile.extractall without path validation enables crafted tar.gz files containing .. or absolute paths to escape the intended extractio...

Malicious code in @wagni_bot/ethereum-wallet (npm)

The npm package @wagnibot/ethereum-wallet is a supply-chain credential stealer disguised as a an Ethereum wallet library SDK. It is one member of a coordinated campaign of 25 crypto/web3 typosquat packages published under the single npm scope @wagnibot on 2026-07-09. Each package declares a...

Malicious code in antsrcsrctest (npm)

The npm package antsrcsrctest masquerades as a "simple date formatting utility" index.js exports a harmless formatDate function that is never referenced by the malicious code but is a credential-harvesting and cloud-metadata-stealing reconnaissance tool. It declares a preinstall: node preinstall....

CVE-2026-14967

BBOT's githubworkflows module could be induced to write a downloaded artifact outside its configured output directory: its path-containment check did not resolve .., so a crafted CODEREPOSITORY URL could traverse out of the intended folder. The write is bounded to two directory levels above the...

CVE-2026-14967

BBOT’s github_workflows module has a path-traversal vulnerability: the path-containment check fails to resolve ‘..’, allowing a crafted CODE_REPOSITORY URL to traverse out of the configured output directory. The write is bounded to two directory levels above the output location and the target is ...

CVE-2026-14967 Path traversal in github_workflows allows writing artifacts outside output directory

BBOT's githubworkflows module could be induced to write a downloaded artifact outside its configured output directory: its path-containment check did not resolve .., so a crafted CODEREPOSITORY URL could traverse out of the intended folder. The write is bounded to two directory levels above the...

CVE-2026-58426

Gitea Actions Artifacts V4 signed URL HMAC ambiguity allows cross-repository artifact read and cross-task upload-state write...

CVE-2026-58426

The CVE-2026-58426 affects Gitea, specifically the Actions Artifacts V4 signed URL mechanism, where an HMAC ambiguity enables cross-repository artifact read and cross-task upload-state write. The vulnerability stems from how signed URLs are validated, allowing unauthorized access across repositor...

CVE-2026-58426 Gitea Actions Artifacts V4 signed URL HMAC ambiguity allows cross-repository artifact read and cross-task upload-state write

Gitea Actions Artifacts V4 signed URL HMAC ambiguity allows cross-repository artifact read and cross-task upload-state write...

CVE-2026-58426 Gitea Actions Artifacts V4 signed URL HMAC ambiguity allows cross-repository artifact read and cross-task upload-state write

Gitea Actions Artifacts V4 signed URL HMAC ambiguity allows cross-repository artifact read and cross-task upload-state write...

CVE-2026-58426

Gitea Actions Artifacts V4 signed URL HMAC ambiguity allows cross-repository artifact read and cross-task upload-state write...

PT-2026-55658

Name of the Vulnerable Software and Affected Versions Gitea affected versions not specified Description An HMAC ambiguity in Gitea Actions Artifacts V4 signed URLs allows for cross-repository artifact reading and cross-task upload-state writing. HMAC Hash-based Message Authentication Code is a...

CVE-2026-9563

In Eclipse Parsson published Maven Central artifacts before version 1.1.8, the JSON parser did not enforce a default maximum on the number of characters consumed while parsing a single JSON document. Applications that parse attacker- controlled JSON can be forced to consume excessive CPU and memo...

jq: embedded NUL in jq import paths causes local redaction-policy bypass and preserves sensitive fields in published artifacts

A flaw was found in jq, a command line JSON processor. Embedded NUL bytes in import paths are truncated during module and data-file lookup, creating a mismatch between the intended import string and the actual file path opened. This issue allows an attacker who can supply a crafted script to acce...

Astra Linux – Vulnerability in docker.io-app

Docker Compose relies on the path information embedded in remote OCI Compose artifacts. When a layer includes the annotations com.dockercompose.extends or com.dockercompose.envfile, Compose incorporates the value provided by the attacker from com.dockercompose.file/com.dockercompose.envfile into...

CVE-2026-56370 ImageMagick - Out-of-bounds Access in ConnectedComponentsImage via connected-components Artifact

ImageMagick before 7.1.2-19 contains an out-of-bounds access vulnerability in ConnectedComponentsImage when processing connected-components artifacts with invalid indices. Attackers can trigger access violations by specifying malformed connected-components definitions via CLI, causing denial of...

CVE-2026-56370 ImageMagick - Out-of-bounds Access in ConnectedComponentsImage via connected-components Artifact

ImageMagick before 7.1.2-19 contains an out-of-bounds access vulnerability in ConnectedComponentsImage when processing connected-components artifacts with invalid indices. Attackers can trigger access violations by specifying malformed connected-components definitions via CLI, causing denial of...

PYSEC-2026-242

Langflow is a tool for building and deploying AI-powered agents and workflows. Prior to 1.9.0, Langflow's /api/v1/monitor router exposes 7 endpoints that perform read, write, and delete operations on user-owned resources — messages, sessions, build artifacts, and LLM transaction logs — without...

CVE-2026-47155

vLLM is an inference and serving engine for large language models LLMs. Prior to 0.22.0, vLLM's revision pinning controls do not consistently apply to all artifacts loaded for a model. A deployment that supplies --revision or --code-revision can still load dynamic code, GGUF files, image...

PYSEC-2026-2301

vLLM is an inference and serving engine for large language models LLMs. Prior to 0.22.0, vLLM's revision pinning controls do not consistently apply to all artifacts loaded for a model. A deployment that supplies --revision or --code-revision can still load dynamic code, GGUF files, image...