CVE-2026-75130

Context7 through 2.1.2 contains a prompt injection vulnerability that allows attackers to execute malicious instructions in connected AI coding agents by injecting unsanitized content through the Custom AI Instructions feature served via the MCP server. Attackers can poison the custom instruction...

CVE-2026-74038

Wazuh 4.0.0 before 4.14.6 contains a path traversal vulnerability that allows unauthenticated remote attackers to cause denial of service by enrolling an agent with a dot-sequence name such as ".." through the enrollment port. Attackers exploit insufficient validation in OSIsValidName and unsafe...

EUVD-2026-61074

Wazuh 4.0.0 before 4.14.6 contains a path traversal vulnerability that allows unauthenticated remote attackers to cause denial of service by enrolling an agent with a dot-sequence name such as ".." through the enrollment port. Attackers exploit insufficient validation in OSIsValidName and unsafe...

CVE-2026-74038 Wazuh 4.0.0 < 4.14.6 Path Traversal DoS via Agent Enrollment

Wazuh 4.0.0 before 4.14.6 contains a path traversal vulnerability that allows unauthenticated remote attackers to cause denial of service by enrolling an agent with a dot-sequence name such as ".." through the enrollment port. Attackers exploit insufficient validation in OSIsValidName and unsafe...

CVE-2026-75911

CodeWhale versions before 0.8.64 fail to properly validate the allowshell configuration parameter from project config files, allowing attackers to enable arbitrary shell command execution by committing a malicious .codewhale/config.toml file to a repository. When a user clones and opens the...

CVE-2026-75898

RAGFlow before 0.26.3 contains a server-side request forgery vulnerability in the agent workflow "Invoke" component agent/component/invoke.py. The component builds an outbound request URL from canvas configuration and runtime template variables and passes it to requests.get, requests.post, or...

CVE-2026-75898 RAGFlow < 0.26.3 - Server-Side Request Forgery via Agent Invoke Component

RAGFlow before 0.26.3 contains a server-side request forgery vulnerability in the agent workflow "Invoke" component agent/component/invoke.py. The component builds an outbound request URL from canvas configuration and runtime template variables and passes it to requests.get, requests.post, or...

EUVD-2026-60824

RAGFlow before 0.26.3 contains a server-side request forgery vulnerability in the agent workflow "Invoke" component agent/component/invoke.py. The component builds an outbound request URL from canvas configuration and runtime template variables and passes it to requests.get, requests.post, or...

CVE-2026-75898 RAGFlow < 0.26.3 - Server-Side Request Forgery via Agent Invoke Component

RAGFlow before 0.26.3 contains a server-side request forgery vulnerability in the agent workflow "Invoke" component agent/component/invoke.py. The component builds an outbound request URL from canvas configuration and runtime template variables and passes it to requests.get, requests.post, or...

CVE-2026-75898

RAGFlow before 0.26.3 lacks URL validation in the agent Invoke component (agent/component/invoke.py), allowing low-privilege users to force the server to fetch internal metadata endpoints and return responses. Patch is available in v0.26.3. Exploitation details are not provided in the documents.

CVE-2026-59781

When Zabbix Agent was installed on Windows into a custom installation directory, the installer did not verify whether the selected directory had secure access permissions. If the target directory allowed unauthorized users to modify its contents, an attacker could place a malicious DLL that could...

AI "Mind Viruses" Can Spread Between Agents Through Persistent Prompt Files

Security researchers at Anthropic and Switzerland's EPFL have demonstrated that self-propagating payloads can spread from one artificial intelligence AI agent to the next through the editable system prompt files that autonomous agent harnesses use to carry state between sessions. The work, releas...

CVE-2026-59781 Improper validation of custom installation directories on Windows could allow installation into locations with unsafe permissions, increasing the risk of DLL sideloading.

When Zabbix Agent was installed on Windows into a custom installation directory, the installer did not verify whether the selected directory had secure access permissions. If the target directory allowed unauthorized users to modify its contents, an attacker could place a malicious DLL that could...

CVE-2026-59781

When Zabbix Agent was installed on Windows into a custom installation directory, the installer did not verify whether the selected directory had secure access permissions. If the target directory allowed unauthorized users to modify its contents, an attacker could place a malicious DLL that could...

EUVD-2026-60742

When Zabbix Agent was installed on Windows into a custom installation directory, the installer did not verify whether the selected directory had secure access permissions. If the target directory allowed unauthorized users to modify its contents, an attacker could place a malicious DLL that could...

The vulnerability of the Directum Web Agent component of the Directum RX system, which arises due to insufficient validation of input data, allows a perpetrator to execute arbitrary code.

The vulnerability of the Directum Web Agent component of the Directum RX system exists due to insufficient validation of input data. Exploiting this vulnerability could allow a malicious actor to execute arbitrary code using a specially crafted file...

PraisonAI AgentOS - Information Disclosure

PraisonAI's AgentOS FastAPI application server exposes an unauthenticated GET /api/agents endpoint that lists every registered agent's name, role and the opening of its instructions system prompt. No authentication is enforced on the route, allowing a remote attacker to enumerate agent...

PT-2026-77139

RAGFlow before 0.26.3 contains a server-side request forgery vulnerability in the agent workflow "Invoke" component agent/component/invoke.py. The component builds an outbound request URL from canvas configuration and runtime template variables and passes it to requests.get, requests.post, or...

CVE-2026-75482

SWE-agent's trajectory inspector sweagent inspector, confirmed in v1.1.0, is an HTTP server that joins request paths to the trajectory directory in its /trajectory/ handler without rejecting parent-directory '..' references, bypassing the built-in path sanitization. The server binds all interface...

CVE-2026-75482 SWE-agent Trajectory Inspector Path Traversal File Disclosure

SWE-agent's trajectory inspector sweagent inspector, confirmed in v1.1.0, is an HTTP server that joins request paths to the trajectory directory in its /trajectory/ handler without rejecting parent-directory '..' references, bypassing the built-in path sanitization. The server binds all interface...