CVE-2025-67511
CriticalCVSS 9.6Exploitation Probability (EPSS)
High risk80th percentile - higher than 80% of all known CVEs
Summary
Cybersecurity AI (CAI) to otwarte oprogramowanie do budowania i wdrażania automatyzacji opartej na sztucznej inteligencji. Wersje 0.5.9 i wcześniejsze są podatne na wstrzyknięcie poleceń przez funkcję run_ssh_command_with_credentials(), która jest dostępna dla agentów AI, ponieważ wartości nazwy użytkownika, hosta i portu są podatne na wstrzyknięcia.
Risk Assessment
Wykorzystanie tej podatności może prowadzić do nieautoryzowanego dostępu do systemów, co stwarza poważne zagrożenie dla bezpieczeństwa organizacji. Brak dostępnej poprawki zwiększa ryzyko eksploatacji tej luki.
Recommendation
Zaleca się aktualizację do najnowszej wersji CAI, gdy poprawka zostanie wydana, oraz monitorowanie systemów pod kątem nieautoryzowanych działań. Należy również rozważyć ograniczenie dostępu do funkcji run_ssh_command_with_credentials.
Related vulnerabilities
- CVE-2026-108860Critical
BotSharp through 5.2.0 contains an authentication bypass vulnerability that allows unauthenticated remote attackers to forge bearer tokens using the hard-coded Jwt:Key in WebStarter appsettings.json. Attackers can sign tokens with the committed HMAC secret and fixed botsharp issuer and audience to impersonate any known user, including administrators, on Authorize-protected API routes.
- CVE-2026-108753Critical
Agnaistic agnai through 1.0.555 contains a hard-coded credentials vulnerability in self-host.docker-compose.yml, which sets a fixed admin password and public JWT secret. Unauthenticated attackers can log in as admin or sign their own JWT with admin: true to impersonate users, reset passwords, and change server configuration.
- CVE-2026-108576Critical
A vulnerability was found in TOZED X300 up to 6.01.3. This vulnerability affects the function process_ping of the component IPPingDiagnostics Handler. The manipulation of the argument Host results in os command injection. The attack can be launched remotely. The vendor was contacted early about this disclosure but did not respond in any way.
- CVE-2026-86717Critical
The Insurify WordPress plugin through 1.0 does not have authorisation and nonce checks on one of its AJAX actions, allowing unauthenticated users to delete arbitrary WordPress options, which can take the site offline and strip every user of their role.
- CVE-2026-86706Critical
The Quick quotes WordPress plugin through 1.0.0 does not perform any capability or nonce check on one of its AJAX actions and lets the caller choose which option is written, allowing unauthenticated users to alter arbitrary site settings and to make the site unavailable.
- CVE-2026-85121Critical
The Insurify WordPress plugin through 1.0 does not have authorisation and nonce checks on one of its AJAX actions, allowing unauthenticated users to create and overwrite arbitrary WordPress options with request data, which can take the site offline and deactivate all of its Insurify WordPress plugin through 1.0.
- CVE-2026-85118Critical
The AI Content Generator Marketing WordPress plugin through 1.0.0 does not enforce a nonce or capability check on some of its AJAX actions, allowing unauthenticated users to update and delete arbitrary WordPress options, which can be used to gain administrator access to the site.
- CVE-2026-84737Critical
The Freeton WP WordPress plugin through 1.0.0 does not correctly validate the activation code when authenticating a user, allowing unauthenticated attackers to log in as any user whose email address they know, including administrators.
- CVE-2026-84734Critical
The Mindstien Quick Login WordPress plugin through 1.0 does not correctly validate a value supplied in the request against the visitor's own session before authenticating them, allowing unauthenticated attackers to obtain a session as the administrator account the Mindstien Quick Login WordPress plugin through 1.0 is configured with.
- CVE-2026-84254Critical
The click5 CRM add-on to Contact Form 7 WordPress plugin through 1.0.4 does not have authorisation and CSRF checks when updating options via a REST endpoint, and does not ensure that the option to be updated belongs to the click5 CRM add-on to Contact Form 7 WordPress plugin through 1.0.4. As a result, unauthenticated attackers could change arbitrary blog options, allowing them to create a new administrator account and take over the site.
Original NVD description (English source)
Cybersecurity AI (CAI) is an open-source framework for building and deploying AI-powered offensive and defensive automation. Versions 0.5.9 and below are vulnerable to Command Injection through the run_ssh_command_with_credentials() function, which is available to AI agents. Only password and command inputs are escaped in run_ssh_command_with_credentials to prevent shell injection; while username, host and port values are injectable. This issue does not have a fix at the time of publication.
Vulnerability data from NVD (NIST) · CISA KEV · EPSS

