CVE-2026-76072: Continue CLI through 1.5.47 Incomplete Destructive Command Denylist in Headless and Auto Mode

Published Aug 24, 2026
·
Updated

The Continue CLI applies an incomplete denylist as its only barrier to destructive shell commands when running unattended. In headless mode and auto mode the default policy in extensions/cli/src/permissions/defaultPolicies.ts grants the Bash tool the allow permission, and permissionChecker.ts hard-blocks a command only when the terminal-security evaluator returns a disabled verdict, so isCriticalCommand in packages/terminal-security/src/evaluateTerminalCommandSecurity.ts is the sole control. Its dangerous-path test matches only /, /, ~, ~/, /usr, /etc, /bin and /sbin and their prefixes, so a recursive forced removal of /home, /root, /var, /opt or /srv is not disabled. The command line is parsed with shell-quote, which reduces $HOME to an empty token, so rm -rf $HOME also fails the dangerous-path test while the shell re-expands the variable when the command is spawned. find with -delete is rated high risk rather than disabled, and shred, wipefs, truncate and pkexec are not handled. Because the agent autonomously reads content it does not control, including fetched web pages, repository files and issue text, an indirect prompt injection in that content can cause an unattended run to destroy the invoking user's data.

Affected Software

1 affected component
Continue CLI<=1.5.47

Event History

Aug 24, 2026
CVE Published
via MITRE·05:55 PM
Data Sourced
via MITRE·05:55 PM
DescriptionSeverityWeakness

Frequently Asked Questions

1

Which deployments are exposed to this issue?

Deployments running the Continue CLI in headless mode or auto mode are exposed because the default policy allows the Bash tool in those unattended modes. Interactive use is not identified as affected by the provided information.

2

What does an attacker need to exploit it?

An attacker needs to get malicious instructions into content the agent reads, such as a fetched web page, repository file, or issue text. The unattended agent can then be induced through indirect prompt injection to run destructive commands that the denylist does not disable.

3

Which destructive commands can bypass the terminal-security control?

Examples include recursive forced removal targeting /home, /root, /var, /opt, or /srv, as well as rm -rf $HOME because $HOME is parsed as an empty token before the spawned shell expands it. The control also does not handle shred, wipefs, truncate, or pkexec, while find with -delete is classified as high risk rather than disabled.

4

What can be done if patching is not immediately possible?

Avoid running the CLI unattended in headless or auto mode with the default Bash allow policy. In particular, do not allow autonomous runs to consume untrusted fetched pages, repository content, or issue text while Bash execution remains permitted.

Contact

SecAlerts Pty Ltd.
132 Wickham Terrace
Fortitude Valley,
QLD 4006, Australia
info@secalerts.co
By using SecAlerts services, you agree to our services end-user license agreement. This website is safeguarded by reCAPTCHA and governed by the Google Privacy Policy and Terms of Service. All names, logos, and brands of products are owned by their respective owners, and any usage of these names, logos, and brands for identification purposes only does not imply endorsement. If you possess any content that requires removal, please get in touch with us.
© 2026 SecAlerts Pty Ltd.
ABN: 70 645 966 203, ACN: 645 966 203