CVE Watch27 de out. de 2026

CVE-2026-113726

virtualenv is a tool for creating isolated virtual python environments. Prior to 21.13.0, the bash activate script that zsh also sources lea

CVSS

7.8

Alta

EPSS

0.3%

p28

KEV

—

Exploit Today

0-100

Publicado: 20 de out. de 2026 · Última mod.: 27 de out. de 2026 · CWE-78 · CWE-116

EPSS · 30d

0.2%EPSS · 30 dias0.3%

2026-10-212026-11-07

Descrição técnica

virtualenv is a tool for creating isolated virtual python environments. Prior to 21.13.0, the bash activate script that zsh also sources leaves the environment name unescaped; from 21.13.0 until 21.14.4, it escapes backslashes, dollar signs, and backticks only when BASH_VERSION is set, so zsh receives the raw value. The script assigns that value to PS1. In zsh with PROMPT_SUBST, an attacker-influenced environment directory, --prompt, or VIRTUALENV_PROMPT value containing command, parameter, or arithmetic expansion syntax can therefore execute commands in the user's shell each time the prompt is drawn. A percent sign is separately interpreted as a zsh prompt escape but does not by itself execute commands. Plain zsh with PROMPT_SUBST disabled and bash are not affected by this mechanism. This issue is fixed in version 21.14.4.

Referências oficiais

CVEs relacionados

CVECVSSEPSSKEVRExploitTítuloMod.

CVE-2026-1235402.0 BAI

—

———OpenAM before 16.1.4 contains an incomplete output encoding vulnerability that allows a SAML 1.x identity provider to alter attribute values by injecting unencoded content into the service provider auto-submitted form. Attackers controlling a trusted identity provider can craft assertion attribute names or values that reach the target application in modified form, undermining the integrity of received attributes.10h CVE-2026-174089.8 CRÍ

60.7%

——18IBM API Connect 10.0.8.0 through 10.0.8.12 and 12.1.0.0 through 12.1.1.2 could allow a remote attacker to execute arbitrary code due to improper neutralization of special elements in an OS command.1d CVE-2026-174029.9 CRÍ

79.9%

——24IBM API Connect 10.0.8.0 through 10.0.8.12 and 12.1.0.0 through 12.1.1.2 could allow a remote authenticated attacker to execute arbitrary commands due to OS command injection. User-supplied input is not properly neutralized before it is passed to the operating system, which could allow commands to run on the underlying host.1d CVE-2026-1030225.3 MÉD

41.1%

——12Urwid is a console user interface library for Python. Prior to 4.0.7, handle_short_request() in urwid/display/web.py performs unguarded character encoding on POST input and unguarded UTF-8 decoding on polling update socket output. A remote attacker who obtains a valid X-Urwid-ID session token can submit non-ASCII POST data that raises UnicodeEncodeError, while non-UTF-8 polling output raises UnicodeDecodeError, causing the CGI handler to return an error and terminate request processing. The POST exception path also leaves an opened file descriptor unclosed for the CGI process lifetime, so repeated requests can cause denial of service for the web display interface. This issue is fixed in version 4.0.7.2d CVE-2026-1026106.4 MÉD

21.8%

——7fast-agent is a framework for building, running, and evaluating AI agents with Model Context Protocol support. Prior to 0.7.21, fast-agent serve uses HTTP transport bound to 0.0.0.0 on port 8000 without authentication, allowing any network client that can reach the /mcp endpoint to invoke the exposed agent tool. When the server is started with --shell, the agent can call the local execute tool implemented by tools/local_shell_executor.py, so an unauthenticated remote client can cause operating-system commands to run as the server user; without --shell, the client can still consume the operator's model and configured MCP tools. The defaults are defined in cli/commands/serve.py and mcp/server/agent_server.py, and the non-interactive server permission prompt does not prevent the described access. This issue is fixed in version 0.7.21.2d CVE-2026-102530—

36.2%

——11JupyterLab Desktop is the cross-platform desktop application for JupyterLab. Prior to 4.6.2-1, remote-server URLs were inserted into the server-selection dialog and Welcome view without safe encoding. An attacker who convinces a user to submit a crafted URL can execute JavaScript in the Welcome renderer and abuse its privileged IPC bridge to execute arbitrary code with the user’s privileges. The Welcome view also unsafely renders RSS news-feed titles; exploitation through this separate input path requires control over content delivered through the application’s Jupyter Blog feed or its cached feed data, rather than submission of a remote-server URL. This issue affects all supported platforms and is fixed in version 4.6.2-1.2d