Meta Muse macOS Client Zero-Day Bypasses Permissions via Config Key

1 reports archived1 independent sourcesupdated 10/8/2026, 3:50:05 AM

Synthesis & Latest Updates

1 Sources Cross-Validated

Security researcher Patrick Wardle disclosed an unpatched zero-day vulnerability in Meta's Muse macOS client, allowing attackers to hijack voice dictation traffic by tampering with the endo_voyager_dictation_endpoint configuration key. The flaw enables non-privileged processes to bypass macOS permission prompts, stealing raw audio and authentication tokens to facilitate prompt injection and unauthorized actions. Meta has not yet issued a security advisory or requested a CVE.

LATEST/Security researcher Patrick Wardle disclosed an unpatched zero-day vulnerability in Meta's Muse macOS client, allowing attackers to hijack voice dictation traffic by tampering with the endo_voyager_dictation_endpoint configuration key. The flaw enables non-privileged processes to bypass macOS permission prompts, stealing raw audio and authentication tokens to facilitate prompt injection and unauthorized actions. Meta has not yet issued a security advisory or requested a CVE.

TIMELINECoverage timeline

Total 1 reports · Latest first

InfoQ 中文T2·Oct 8, 03:50 AM68 pts
Meta Muse macOS Client Zero-Day Bypasses Permissions via Config Key
Original: 苹果筑起的权限高墙，被 Meta AI 助手“借道”绕过
- Non-privileged local processes can silently overwrite critical Muse client configuration keys, hijacking traffic without triggering macOS authorization prompts.
- Attackers can simultaneously steal raw microphone audio and valid authentication tokens, enabling undetected man-in-the-middle attacks.
- With session credentials, attackers can use prompt injection to drive the AI assistant to perform malicious background tasks like file exfiltration.
View Evidence Original Source