Skip to content
Critical OS Command Injection Vulnerability in Ai Sdlc MCP Server Tools

Critical OS Command Injection Vulnerability in Ai Sdlc MCP Server Tools

First seen 28 Oct 2026, 02:21 UTC •
API

Article Content

Browse articles
ThreatCluster AI
ThreatCluster •October 28, 2026 at 04:19 UTC
  • •Critical OS command injection vulnerability in ai-sdlc MCP server tools.
  • •CVSS score of 9.8 indicates high severity and potential for exploitation.
  • •Attackers can execute arbitrary commands due to lack of input validation.

The ai-sdlc MCP server tools, specifically check_issue and check_pr_status, are vulnerable to OS command injection (CWE-78) due to improper input handling. The vulnerability allows attackers to execute arbitrary commands on the host system by exploiting the repo parameter, which is directly interpolated into execSync() calls without sanitization. This critical vulnerability has a CVSS score of 9.8 and affects all versions up to 0.31.2. Attack vectors include shell command separators, subshell expansion, and pipes, enabling attackers to exfiltrate sensitive data or establish reverse shells. The vulnerability was disclosed on October 28, 2026, and is currently. Security professionals are urged to review their use of these tools and implement mitigations as soon as possible.

Start a free Starter trial for enhanced analysis

Ask AI about this cluster

Updated just now How this analysis works

Timeline

2026-10-28
Vulnerability disclosed
The ai-sdlc MCP server tools were found to be vulnerable to OS command injection, allowing arbitrary command execution.
github.com
2026-10-28
Vulnerability advisory published
Vulncheck published an advisory detailing the command injection vulnerability in ai-sdlc MCP server tools.
vulncheck.com

More articles in this cluster (2)

Extracted Entities (9)

1 / 4

Common questions

Which versions are affected?
All versions of ai-sdlc MCP server tools up to 0.31.2 are affected by this vulnerability.
Is there a patch available?
No patch has been released as of now, and the vulnerability remains unaddressed.
What should organizations do to protect themselves?
Organizations should implement input validation and consider alternative methods to execute commands securely until a patch is available.