In its design for automatic terminal command execution, AI Code offers two options: Execute safe commands and execute all commands. The description for the former states that commands determined by the model to be safe will be automatically executed, whereas if the model judges a command to be potentially destructive, it still requires user approval. However, this design is highly susceptible to prompt injection attacks. An attacker can employ a generic template to wrap any malicious command and mislead the model into misclassifying it as a 'safe' command, thereby bypassing the user approval requirement and resulting in arbitrary command execution.
References
| Link | Resource |
|---|---|
| https://github.com/Secsys-FDU/LLM-Tool-Calling-CVEs/issues/2 | Issue Tracking |
| https://marketplace.visualstudio.com/items?itemName=tianguaduizhang.claude-dev-china | Product |
Configurations
History
03 Apr 2026, 16:00
| Type | Values Removed | Values Added |
|---|---|---|
| CPE | cpe:2.3:a:tianguaduizhang:ai_code:*:*:*:*:*:visual_studio_code:*:* |
02 Apr 2026, 19:54
| Type | Values Removed | Values Added |
|---|---|---|
| First Time |
Tianguaduizhang ai Code
Tianguaduizhang |
|
| CPE | cpe:2.3:a:tianguaduizhang:ai_code:*:*:*:*:*:*:*:* | |
| References | () https://github.com/Secsys-FDU/LLM-Tool-Calling-CVEs/issues/2 - Issue Tracking | |
| References | () https://marketplace.visualstudio.com/items?itemName=tianguaduizhang.claude-dev-china - Product |
27 Mar 2026, 20:16
| Type | Values Removed | Values Added |
|---|---|---|
| CWE | CWE-20 | |
| CVSS |
v2 : v3 : |
v2 : unknown
v3 : 9.6 |
27 Mar 2026, 15:16
| Type | Values Removed | Values Added |
|---|---|---|
| New CVE |
Information
Published : 2026-03-27 15:16
Updated : 2026-04-03 16:00
NVD link : CVE-2026-30304
Mitre link : CVE-2026-30304
CVE.ORG link : CVE-2026-30304
JSON object : View
Products Affected
tianguaduizhang
- ai_code
CWE
CWE-20
Improper Input Validation