Understand vulnerabilities. Protect your applications.
Discover which vulnerabilities affect your technologies and how to fix them. Explore clear explanations, affected versions and practical recommendations - from CMS platforms and plugins to application libraries.
Vulnerability archive 72,283
virtualenv: Bash prompt values can change meaning under older compatibility modes
An issue in virtualenv allows caller-supplied prompt or destination values to execute arbitrary commands when an environment is activated under certain shell configurations. Users should recreate environments with patched virtualenv releases or disable prompt modification.
virtualenv: Xonsh activation literals depend on the script decoding codec
A vulnerability in virtualenv allows caller-supplied prompt text to become executable code in xonsh activation scripts under legacy multibyte encodings. When an attacker influences the prompt or destination and a user activates the environment, arbitrary commands may execute with the user's privileges.
MLflow FileStore Path Traversal and Arbitrary File Write Vulnerability
A path traversal vulnerability in MLflow FileStore allows unauthenticated attackers to write arbitrary files containing metadata outside the intended trace directory by sending crafted trace IDs to the traces API endpoint.
MSAL Node authorization code flow may disclose parameters to an attacker-controlled endpoint
An origin validation error in Microsoft Authentication Library (MSAL) for Node.js allows unauthorized attackers to disclose information over a network during the authorization code flow.
undici vulnerable to Denial of Service via unhandled error on excessive content-encodings
A remote server can trigger a denial of service in undici by sending a response with excessive content codings and abruptly closing the connection. This leads to an unhandled error on an orphaned stream, terminating the Node.js process.
multer vulnerable to Denial of Service via a pending asynchronous fileFilter after request teardown
A denial-of-service vulnerability in multer allows an unauthenticated remote attacker to crash the Node.js process via a single request when an asynchronous file filter is configured.
Happy DOM: Incomplete fix for CVE-2026-33943: unsanitized import alias leads to host code execution
A vulnerability in Happy DOM allows untrusted ESM import aliases to be converted into executable JavaScript, enabling code execution on the host when script evaluation is enabled.
Electrum: Malicious server can replace a just-broadcast tx and try to trick user into RBF-ing it
A malicious Electrum server can intercept an unconfirmed transaction and display an invalid replacement. If a user manually bumps fees via RBF, the client signs the transaction and sends funds to the attacker.
nodejs-whisper vulnerable to OS command injection via crafted audio file paths
An OS command injection vulnerability in nodejs-whisper allows attackers to execute arbitrary system commands with Node.js process privileges by supplying crafted audio file paths to convertToWavType().
opentelemetry-java: Quadratic CPU consumption in W3C tracestate parsing
Python Social Auth: OpenID Connect nonce is not bound to the browser session, so a stolen authorization code logs the attacker in as the victim
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, the OpenAI-compatible server configures Starlette CORSMiddleware with allowed_origins set to a wildcard by default.
Autobahn Python: permessage-brotli inflates a whole frame before maxMessagePayloadSize is checked (incomplete fix of GHSA-hxp9-w8x3-p566)
ONNX: Null Pointer Dereference in ScanInferenceFunction Due to Missing Required Attribute
filelock: SoftReadWriteLock follows a lock tree swapped in after its first acquire and writes lock records into directories of the user running the lock
ONNX: Out-of-Bounds Read in RNN/GRU/LSTM Shape Inference (Old Opsets, Missing Rank Check)
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, vllm/tokenizers/hf.py computes max_token_id with an off-by-one upper bound, allowing a pre-tokenized prompt ID equal
vLLM DeepSeek-OCR image processing vulnerable to denial of service
vLLM: Phi-4 multimodal audio encoder materializes and retains full attention matrices: a single max-length audio OOM-crashes the vLLM engine
vLLM is an inference and serving engine for large language models. From 0.19.0 until 0.32.0, authenticated callers can supply a request chat template when no server-side --chat-template is configured,
LangGraph CLI prints the registry push token in `langgraph deploy --verbose` output
vLLM: Responses API bypasses trust_request_chat_template for chat_template_kwargs
ONNX: Denial of service via unchecked expansion of packed sub-byte tensors in ONNX Python APIs
Know which vulnerabilities affect your application
With the DockRay integration and software inventory enabled, DockRay checks reported component versions against known vulnerabilities and notifies you about confirmed matches. Coverage depends on the supported technology and available advisory data.