CVE-2026-12957

Amazon Q language server executes trusted-workspace configuration

Amazon Web Services · Language Servers for AWS / Amazon Q Developer

First documented disclosure · Review checkpoint

Summary

Improper trust-boundary enforcement permits commands in a crafted project configuration to run.

Severity and exploitation evidence

High · 8.5/10 · CVSS 4.0. Severity source.

Exploitation not established: The reviewed sources establish a disclosed weakness or researcher demonstration, not malicious exploitation in the wild.

Technical severity is not a measure of observed attacks or of risk to your deployment. Attempts, demonstrations, and successful compromise retain the qualifications in the source.

How the weakness works

Configuration processing crosses the boundary into command execution.

Conditions required

The local user opens a malicious workspace and confirms workspace trust.

Potential impact

Arbitrary code execution with the language-server process’s permissions.

Affected versions and configurations

Language Servers for AWS before 1.65.0.

Fix and mitigation status

1.65.0 fixes this CVE; AWS recommends 1.69.0 and corresponding plugin updates to cover both June bulletin issues.

Practical next steps

Upgrade the language server and bundled IDE plugin; restrict untrusted workspaces.

Operating lesson

Workspace trust is a consequential authorization, even when the user only intends to inspect a project.

Evidence and limits

Distinct from the symlink CVE-2026-12958 in the same bulletin. Do not apply the 1.69.0 affected range of that flaw to this CVE.

Sources