🚨 CVE-2026-56266
Crawl4AI before 0.8.7 contains a server-side request forgery vulnerability in the /crawl, /crawl/stream, /md, and /llm endpoints that fetch arbitrary user-supplied URLs without validation. Unauthenticated attackers can bypass the internal-address blocklist using IPv6-mapped IPv4 addresses to reach internal services and cloud metadata endpoints.

🎖@cveNotify

🚨 CVE-2026-56268
Flowise before 3.1.2 contains an information disclosure vulnerability in the /api/v1/chatflows/apikey/:apikey endpoint. When the keyonly query parameter is omitted (the default), the endpoint returns not only the chatflows bound to the supplied API key but also all chatflows across every workspace that have no API key assigned, because the underlying query lacks any workspace filter. An attacker with a valid API key for one workspace can therefore retrieve the full ChatFlow configuration (including flowData with system prompts and node configurations, chatbotConfig, apiConfig, and credential IDs) of unprotected chatflows belonging to other workspaces.

🎖@cveNotify

🚨 CVE-2026-56280
Cap-go before 12.128.2 contains a privilege inversion vulnerability in GET /build/logs/:jobId that allows read-only API key holders to cancel running native builds. The endpoint registers an abort listener on the SSE stream that unconditionally invokes cancelBuildOnDisconnect() using the privileged server-side BUILDER_API_KEY when clients disconnect, bypassing the app.build_native permission check required by the explicit POST /build/cancel/:jobId endpoint. Attackers with read-only API keys can repeatedly disrupt native build operations and CI/CD workflows by opening the log stream and dropping the connection.

🎖@cveNotify

🚨 CVE-2026-56306
Capgo before 12.128.2 contains a weak parsing vulnerability in the x-limited-key-id header that allows attackers to bypass subkey enforcement by submitting malformed values, zero, or duplicate headers that result in NaN or falsy values. Remote attackers can manipulate the x-limited-key-id header to disable limited key scoping and execute requests using the main API key context instead of restricted subkey permissions.

🎖@cveNotify

🚨 CVE-2026-56311
Capgo before 12.128.2 contains an authorization bypass vulnerability in the public.get_current_plan_max_org RPC function that allows unauthenticated attackers to retrieve arbitrary organization plan limits. Attackers can call the RPC endpoint with any organization UUID using only the public Supabase key to disclose billing information including MAU, bandwidth, storage, and build time limits for any organization.

🎖@cveNotify

🚨 CVE-2026-56314
Capgo before 12.128.12 fails to filter deleted app versions when joining channels during /updates resolution, allowing deleted bundles to remain selectable. Attackers can continue deploying deleted bundles to devices by exploiting the missing app_versions.deleted filter in channel version joins.

🎖@cveNotify

🚨 CVE-2026-56321
Capgo (backend Supabase edge functions) before 12.128.2 does not apply the global authentication middleware to the GET /private/role_bindings/:org_id endpoint, unlike the POST and DELETE role_bindings routes, so unauthenticated requests reach the handler instead of being rejected at the middleware layer. The handler still performs its own authorization check and returns Unauthorized, so no direct data exposure occurs; the flaw is inconsistent authentication enforcement across HTTP methods that could enable authorization bypass if the handler logic changes.

🎖@cveNotify

🚨 CVE-2026-56323
Capgo before 12.128.2 contains an information disclosure vulnerability in the /functions/v1/channel_self endpoint that allows unauthenticated attackers to enumerate non-public channel names and determine app existence and subscription status. Remote attackers can send GET requests with arbitrary app_id parameters to disclose internal rollout channels, enumerate valid applications across tenants, and leak billing status without authentication or device binding.

🎖@cveNotify

🚨 CVE-2026-56324
Capgo before 12.128.2 contains a rate limit bypass vulnerability in the channel_self endpoint that allows attackers to circumvent rate limiting by rotating the user-controlled device_id parameter. Attackers can send multiple requests per second by changing device_id values to flood the channel_devices table and cause database exhaustion.

🎖@cveNotify

🚨 CVE-2026-56326
Nuxt versions 4.0.0 before 4.4.7 and 3.x before 3.21.7 contain a server-side open redirect vulnerability in navigateTo that fails to properly validate path-normalized payloads like /..//evil.com and /.//evil.com. Attackers can bypass external-host checks using path-normalization techniques to redirect users to attacker-controlled sites via the Location header or meta-refresh, enabling phishing and OAuth authorization-code theft.

🎖@cveNotify

🚨 CVE-2026-56348
n8n before 2.20.0 contains a credential exfiltration vulnerability in the POST /rest/dynamic-node-parameters/options endpoint that allows authenticated users to bypass Allowed HTTP Request Domains restrictions. Attackers with credential access can cause the n8n server to issue HTTP requests with credentials to unauthorized hosts, exfiltrating sensitive authentication data.

🎖@cveNotify

🚨 CVE-2026-56697
Nuxt versions 4.0.0 before 4.4.7 and 3.x before 3.21.7 accept protocol-relative paths such as //evil.com in the reloadNuxtApp function; these pass the script-protocol check but resolve to a cross-origin URL against the current page protocol. Attackers can inject paths like //evil.com to redirect users to attacker-controlled hosts, enabling phishing and OAuth authorization-code theft.

🎖@cveNotify

🚨 CVE-2026-56698
Nuxt versions 4.0.0 before 4.4.7 and 3.x before 3.21.7 fail to validate script-capable URLs in the navigateTo open option, allowing client-side script execution. Attackers can supply javascript: URLs through the open parameter to execute arbitrary scripts in the application's origin when user-controlled input is passed to navigateTo.

🎖@cveNotify

🚨 CVE-2026-48931
A flaw in Node.js HTTP Agent can cause a client to accept as valid a response that is send before the client has sent the request.

This vulnerability affects all supported release lines: **Node.js 22**, **Node.js 24**, and **Node.js 26**.

🎖@cveNotify

🚨 CVE-2026-41523
vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.0, an assert-based security check in vLLM's activation function loading allows any unauthenticated attacker to achieve arbitrary code execution on the server by publishing a malicious HuggingFace model, when vLLM runs in Python optimized mode (python -O or PYTHONOPTIMIZE=1). This vulnerability is fixed in 0.22.0.

🎖@cveNotify

🚨 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 processors, retrieval side weights, or same-repository subfolder weights/config from an unpinned/default revision. This is a supply-chain integrity issue for pinned vLLM deployments. Operators can believe they are serving a reviewed model revision while vLLM resolves behavior-affecting nested or sibling artifacts outside that reviewed revision. This vulnerability is fixed in 0.22.0.

🎖@cveNotify

🚨 CVE-2026-48746
vLLM is an inference and serving engine for large language models (LLMs). From 0.3.0 until 0.22.0, a vulnerability in ASGI web servers and starlette's trust on those web servers enables an authentication bypass of the OpenAI API AuthenticationMiddleware. It allows to use the API without providing the configured VLLM_API_KEY or --api-key. This vulnerability is fixed in 0.22.0.

🎖@cveNotify

🚨 CVE-2026-53923
vLLM is an inference and serving engine for large language models (LLMs). From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels (csrc/quantization/gguf/gguf_kernel.cu) causes partial tensor processing. The output tensor is allocated at full size via torch::empty (uninitialized memory), but the dequantize CUDA kernel processes only a truncated number of elements. The unfilled portion of the output tensor retains whatever was previously in GPU memory. In multi-tenant inference deployments, this residual GPU memory may contain tensor data from other users' inference requests, constituting information disclosure. This vulnerability is fixed in 0.23.1rc0.

🎖@cveNotify

🚨 CVE-2026-54232
vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.1, the vLLM Dockerfile is vulnerable to a dependency confusion attack through the flashinfer-jit-cache package. The package is installed from a custom index (flashinfer.ai/whl/) using --extra-index-url, but the package name was not registered on PyPI, and UV_INDEX_STRATEGY="unsafe-best-match" is set globally. An attacker who registers flashinfer-jit-cache on PyPI with version 0.6.11.post2 can execute arbitrary code as root during the Docker build and backdoor every resulting container image, enabling exfiltration of all user prompts, API credentials, and model data from production vLLM deployments This vulnerability is fixed in 0.22.1.

🎖@cveNotify

🚨 CVE-2026-54233
vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, vLLM's /v1/audio/transcriptions endpoint limits compressed upload size but not decoded PCM output. A 25MB OPUS file expands to ~14.9GB of float32 PCM at decode time. This vulnerability is fixed in 0.23.1rc0.

🎖@cveNotify

🚨 CVE-2026-54235
vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, ll temperature validation gates use comparison operators (<, >), which silently evaluate to False for NaN and for positive Infinity in Python's IEEE 754 float semantics. Both values pass every guard and propagate to GPU sampling kernels, where they produce undefined behavior or CUDA errors that can crash the inference worker. This vulnerability is fixed in 0.23.1rc0.

🎖@cveNotify