Threat briefs

A code injection vulnerability (CVE-2026-46633) exists in Twig versions prior to 3.26.0, where a single quote in the `{% use %}` template name is not properly escaped, allowing arbitrary PHP code execution by bypassing the Twig sandbox.

The js-cookie library is vulnerable to prototype pollution via the `assign()` function when processing JSON-derived objects, enabling an attacker to inject arbitrary cookie attributes by manipulating the `__proto__` property, as demonstrated by CVE-2026-46625.

CVE-2026-47102 describes a privilege escalation vulnerability in LiteLLM versions prior to 1.83.10, where the /user/update endpoint allows users to modify their own user_role, potentially escalating their privileges to proxy_admin.

LiteLLM versions prior to 1.83.14 allow an authenticated internal user to create API keys that bypass role-based access controls, potentially leading to full privilege escalation to proxy_admin due to a lack of verification of specified routes within the user's own permissions (CVE-2026-47101).

Attackers are compromising npm packages to distribute a RAT linked to PolinRider, directly injecting malicious code into the software supply chain.

@hulumi/policies versions before 1.3.2 allowed unrelated compliant-looking evidence to suppress violations for different zones, hostnames, origins, or repositories in the same stack, bypassing Cloudflare and deployment-governance guardrails.

Russh versions up to 0.60.2 are vulnerable to a memory-safety hardening issue due to unchecked `CryptoVec` allocation and growth handling, reachable from local agent inputs and remote SSH traffic, potentially triggering a process abort under constrained memory conditions.

@hulumi/policies versions before 1.3.2 are vulnerable to a critical trust policy bypass where set-qualified operators such as ForAnyValue:StringLike could hide wildcard GitHub Actions OIDC sub conditions from the mandatory guardrail, potentially allowing unauthorized access to AWS resources.

@hulumi/drift versions before 1.3.2 could accept externally supplied execute plans without sufficient provenance checks, allowing unsafe reconciliation input to be treated as trusted; upgrade to version 1.3.2 or later to resolve this vulnerability.

@hulumi/policies versions before 1.3.2 improperly inspect inline and attached IAM policies, potentially allowing admin-equivalent policy paths to bypass the administrator-policy guardrail, resulting in a CIS 1.16 admin policy bypass.

MCP Server Kubernetes versions before 3.6.0 have an access control bypass vulnerability (CVE-2026-46519) where tool access controls are enforced only at the discovery layer, allowing authenticated clients to invoke any Kubernetes tool regardless of configured restrictions, potentially leading to cluster compromise.

This rule detects potential session hijacking or token replay in Microsoft Entra ID, identifying cases where a user signs in and subsequently accesses Microsoft Graph from a different IP address using the same session ID, which may indicate a successful OAuth phishing attack, session hijacking, or token replay attack.

This rule correlates Entra-ID or Microsoft 365 mail successful sign-in events with network security alerts by source address, indicating potential initial access by adversaries triggering network security alerts before accessing cloud resources.

The p3-challenger rust package is vulnerable to transcript malleability and challenge entropy loss, allowing attackers to craft distinct transcripts that produce identical challenges, breaking the binding property of Fiat-Shamir due to partial-chunk aliasing, non-injective squeeze, and high-bit truncation.

KnpLabs knp-snappy versions 1.7.0 and earlier are vulnerable to command injection (CVE-2026-46643) due to an inverted is_executable check, which prevents proper shell escaping of the binary path, potentially leading to command execution if the binary path is attacker-influenced.

IINA before 1.4.3 contains a user-assisted command execution vulnerability that allows remote attackers to execute arbitrary commands by supplying malicious mpv_-prefixed query parameters through the iina://open custom URL scheme handler.

Fission runtime pods were created with the `fission-fetcher` service account, granting namespace-wide `get` access to secrets and configmaps; the runtime pod's automounted token was reachable from inside the user's function container, allowing user-supplied function code to inherit the same Kubernetes API privileges and read any secret or configmap in the function's namespace, far beyond the intended `Function.spec.secrets` allowlist.

The Fission router exposes the `/fission-function/<ns>/<name>` endpoint on its public listener, allowing invocation of any function without an HTTPTrigger, leading to unauthorized function access and potential cross-tenant exploitation; patched in v1.23.0.

The Fission `storagesvc` component exposes unauthenticated CRUD operations on the `/v1/archive` endpoint, allowing any workload within the same Kubernetes cluster to enumerate archive IDs, download archives, upload arbitrary content, and delete archives, leading to potential code and secret exposure and function disruption.

A remote denial-of-service vulnerability (CVE-2026-46545) exists in Nimiq primitives where an unauthenticated peer can send a malicious chunk with an empty key, leading to a panic when `put_raw` attempts to store a value at the root node, causing the node process to abort.

LMDeploy <= 0.12.3 is vulnerable to remote code execution (CVE-2026-46517) because it hardcodes `trust_remote_code=True` when calling `transformers.AutoConfig.from_pretrained()`, allowing a malicious Hugging Face repository to execute arbitrary Python code when loaded without user opt-out.

Open ISES Tickets before version 3.44.2 contains hardcoded MySQL database connection credentials in import_mdb.php, allowing unauthorized database access.

Open ISES Tickets before version 3.44.2 contains hardcoded MySQL database credentials in loader.php, allowing an attacker with access to the source code or the file on a deployed installation to read the username, password, and database name and use them to connect to the database (CVE-2026-48241).

Open ISES Tickets before version 3.44.2 is vulnerable to SQL injection in ajax/statistics.php via the tick_id and f_tick_id POST parameters, allowing authenticated attackers to manipulate SQL queries and potentially read, modify, or destroy database contents.

Open ISES Tickets before version 3.44.2 is vulnerable to SQL injection via the tick_id POST parameter in ajax/reports.php, allowing an authenticated attacker to alter query semantics to read, modify, or destroy database contents.

Open ISES Tickets before version 3.44.2 is vulnerable to SQL injection (CVE-2026-48238) because the id GET parameter in ajax/mobile_main.php is concatenated into the WHERE clause of a SELECT statement without sanitization, allowing authenticated attackers to craft requests that can read, modify, or destroy database contents.

Open ISES Tickets versions before 3.44.2 are vulnerable to SQL injection (CVE-2026-48235) due to unsanitized GPS data, allowing attackers to manipulate responder data by compromising the GPS tracker endpoint.

A commodity BadIIS malware variant is fueling a thriving malware-as-a-service (MaaS) ecosystem for Chinese-speaking cybercrime groups, allowing them to execute malicious SEO fraud, hijack server content, and redirect traffic to illicit sites.

A cross-site scripting (XSS) vulnerability exists in md-fileserver's Markdown rendering logic, where user-supplied Markdown content containing raw HTML, including <script> tags, is processed and injected into the resulting page without sanitization, leading to arbitrary JavaScript execution and potential account takeover.

ConnectWise released a security advisory addressing a vulnerability in ConnectWise Automate versions prior to 2026.5, prompting users to apply the necessary updates.