cbcvebase.
CVE-2025-55319
published 2025-09-12

CVE-2025-55319: Ai command injection in Agentic AI and Visual Studio Code allows an unauthorized attacker to execute code over a network.

PriorityP267critical9.8CVSS 3.1
AVNACLPRNUINSUCHIHAH
EPSS
0.85%
53.9th percentile
Ai command injection in Agentic AI and Visual Studio Code allows an unauthorized attacker to execute code over a network.

Affected

3 ranges
VendorProductVersion rangeFixed in
microsoftvisual_studio_code< 1.104.01.104.0
microsoftvisual_studio_code>= 1.0.0 < 1.104.01.104.0
msrcvisual_studio_code

Detection & IOCsextracted from sources · hover to see the quote

  • CVE-2025-55319 is an AI command injection vulnerability in Agentic AI and Visual Studio Code allowing remote code execution over a network by an unauthorized attacker — monitor for unexpected process spawning or network-initiated code execution originating from VS Code Agentic AI components
  • Microsoft rates exploitation as 'More Likely' for the latest software release — prioritize patching VS Code to v1.104 (released 2025-09-11) and monitor for exploitation attempts against Agentic AI features
  • Fix is available in VS Code v1.104; reference the release notes at the linked URL to identify the patched version for detection baseline
  • ·The affected component is specifically 'Agentic AI and Visual Studio Code' — detections and mitigations should be scoped to environments where VS Code Agentic AI features are enabled or exposed over a network

CVSS provenance

nvdv3.19.8CRITICALCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
vendor_msrc8.8HIGH
Stop checking back — get the weekly exploitation signal.

Every Monday: what got weaponized or added to CISA KEV in the last seven days — each CVE cross-linked to its PoC, Nuclei template, and detection rule. Free, one email a week, unsubscribe in one click.