Threat Heartbeat

New CVEs as they are published, with CVSS score, CISA KEV status, public exploits and the vendor's fix. Worth tracking, but only a small part of what Panop does: mapping what you expose to the internet, testing what an attacker can actually exploit, ranking it by business impact, and helping your team understand and fix it.

Book a demo Request a scan

Feed at a glance

Tracked
13360
Critical
2486
CISA KEV
125
Public PoC
437

Most common types

  • Broken Access Control 1691
  • Remote Code Execution 1609
  • Privilege Escalation 1307
  • Authentication Bypass 1121
  • Denial of Service 799

Published per day, last 14 days

21 Oct Critical Other 3 Nov

Source: public CVE records (NVD, CISA KEV) · last published 3 Nov 2026

Latest CVEs

Newest first · severity as published by the vendor or NVD · page 21 of 891

  • 8.8 /10 High
    CVE-2026-76026 No auth

    Google Chrome - Remote Code Execution

    Google Chrome < 156.0.8078.79 contains a memory corruption vulnerability in Autofill, letting remote attackers execute arbitrary code inside the sandbox via crafted HTML pages, exploit requires no special privileges.

  • 8.8 /10 High
    CVE-2026-76025 No auth

    Google Chrome - Use After Free

    Google Chrome < 156.0.8078.79 contains a use after free caused by improper memory handling in FileSystem, letting remote attackers execute arbitrary code inside the sandbox via crafted HTML pages.

  • 7.5 /10 High
    CVE-2026-76012 Auth required

    NextGEN Gallery - Remote Code Execution

    NextGEN Gallery WordPress plugin <= 4.4.1 contains a remote code execution caused by broken authorization and unsafe file upload in legacy_render function, letting authenticated attackers with custom-level access execute code, exploit requires attacker to be stored author of at least one gallery.

  • 8.5 /10 High
    CVE-2026-75882 Auth required

    ESET Endpoint Encryption & Full Disk Encryption - Broken Access Control

    ESET Endpoint Encryption and ESET Full Disk Encryption contain an improper access control vulnerability caused by insufficient permission checks in a privileged kernel driver interface, letting attackers with standard user permissions send crafted requests.

  • 8.8 /10 High
    CVE-2026-73186 Auth required

    Enable Media Replace - PHP Object Injection

    Enable Media Replace <= 4.2.2 contains a PHP Object Injection caused by unsafe deserialization, letting attackers execute arbitrary code or commands remotely, exploit requires crafted input.

  • 8.8 /10 High
    CVE-2026-65566 No auth

    Meow Apps AI Engine - Cross Site Request Forgery

    Meow Apps AI Engine <= 3.8.8 contains a cross site request forgery caused by lack of CSRF protections, letting attackers force authenticated users to perform unintended actions, exploit requires victim interaction.

  • 7.6 /10 High
    CVE-2026-63414 Auth required

    Shopify App - Broken Access Control

    Shopify App 22.1.0 to 23.0.3 contains a broken access control caused by token-exchange authenticated controllers exposing request-derived shop context instead of verified Shopify ID token or session, letting attackers with a valid session token for one shop access or act on another shop's data, exploit requires token-exchange authentication and reliance on current_shopify_domain or params[:shop].

  • 7.4 /10 High
    CVE-2026-63398

    Coop - Broken Access Control

    Coop prior to 1.0.2 contains a broken access control vulnerability caused by SAML strategy's verify callback resolving users by email without organization binding, letting attackers with control of a SAML IdP signing key perform cross-tenant account takeover, exploit requires matching signing certificate in target organization.

  • 8.7 /10 High
    CVE-2026-63396 No auth

    ROOST Coop - Information Disclosure

    ROOST Coop < 1.0.0 contains an information disclosure caused by unauthenticated access to Query.allOrgs in /api/v1/graphql, letting unauthenticated actors obtain sensitive organization and user data, exploit requires public GraphQL API access.

  • 8.7 /10 High
    CVE-2026-63393 Auth required

    OpenSearch SQL - Remote Code Execution

    OpenSearch SQL 2.x >= 2.8.0.0 < 2.19.6.0 and 3.x < 3.7.0.0 contain an insecure deserialization vulnerability caused by unsafe deserialization of untrusted data in the plugins/sql endpoint, letting remote authenticated users with read/search permissions execute arbitrary code on the server, exploit requires authenticated user with read/search permissions.

  • 8.8 /10 High
    CVE-2026-63378 Auth required

    TAKO AI Agent for Okta - Command Injection

    TAKO AI Agent for Okta < 3.0.1-beta contains a command injection caused by insufficient validation of dynamic getattr calls in script_code field, letting authenticated users execute arbitrary Python code as root, exploit requires valid fctr_session cookie.

  • 9.8 /10 Critical
    CVE-2026-63377 No auth

    TAKO AI Agent for Okta - Authentication Bypass

    TAKO AI Agent for Okta < 3.0.1-beta contains an authentication bypass caused by lack of authentication and token checks in POST /api/auth/setup, letting unauthenticated network attackers create the first admin account, exploit requires initial setup not completed.

  • 8.6 /10 High
    CVE-2026-63346

    Refitter - Remote Code Execution

    Refitter < 2.1.2 contains a code injection vulnerability caused by unescaped quotation marks in header parameter names, letting attackers execute arbitrary code at assembly load via crafted OpenAPI descriptions, exploit requires attacker-controlled OpenAPI input.

  • 8.6 /10 High
    CVE-2026-63345

    Refitter - Code Injection

    Refitter < 2.1.2 contains a code injection caused by unescaped OpenAPI path keys in Refit route attributes, letting attackers execute code via a crafted path in generated assemblies, exploit requires compilation and loading of generated code.

  • 8.6 /10 High
    CVE-2026-63344

    Refitter - Code Injection

    Refitter < 2.1.2 contains a code injection caused by unescaped content-type keys in OpenAPI descriptions, letting attackers execute arbitrary code via generated assemblies at application startup, exploit requires attacker-controlled OpenAPI input.