Paperclip AI flaws let a malicious agent import run commands on the host

Researchers disclosed flaws in Paperclip, an AI agent tool, that let attackers run commands on a user's machine through malicious agent imports. The more serious path, CVE-2026-41679 and scored 10.0, needs no account or user interaction against network-reachable deployments running in authenticated mode with the default registration configuration. A second path, scored 9.6, requires a victim to open an attacker-controlled page while Paperclip runs in its default local trusted mode. A third issue exposed sensitive data and control-plane details through API routes that skipped access checks. The fixes, which add import authorization and hostname validation, ship in the version tagged 2026.416.0.

Check

Update Paperclip to the 2026.416.0 release, and review whether any AI agent tools you run accept agent definitions or imports from untrusted sources without authorization.

Affected

Users of Paperclip AI running network-reachable authenticated deployments or the default local trusted mode (CVE-2026-41679); malicious agent imports or a visited page can lead to host command execution.

Fix

Apply the fixed release, require authorization for agent imports, avoid exposing agent tools to untrusted networks, and treat imported agent definitions as untrusted code that can execute on the host.