mcp-server-semgrep/semgrep-rules/go/lang/security/audit/dangerous-exec-cmd.yaml

MCP Server for Semgrep Integration - static code analysis with AI

rules:
- id: dangerous-exec-cmd
  patterns:
  - pattern-either:
    - patterns:
      - pattern: |
          exec.Cmd {...,Path: $CMD,...}
      - pattern-not: |
          exec.Cmd {...,Path: "...",...}
      - pattern-not-inside: |
          $CMD,$ERR := exec.LookPath("...");
          ...
      - pattern-not-inside: |
          $CMD = "...";
          ...
    - patterns:
      - pattern: |
          exec.Cmd {...,Args: $ARGS,...}
      - pattern-not: |
          exec.Cmd {...,Args: []string{...},...}
      - pattern-not-inside: |
          $ARGS = []string{"...",...};
          ...
      - pattern-not-inside: |
          $CMD = "...";
          ...
          $ARGS = []string{$CMD,...};
          ...
      - pattern-not-inside: |
          $CMD = exec.LookPath("...");
          ...
          $ARGS = []string{$CMD,...};
          ...
    - patterns:
      - pattern: |
          exec.Cmd {...,Args: []string{$CMD,...},...}
      - pattern-not: |
          exec.Cmd {...,Args: []string{"...",...},...}
      - pattern-not-inside: |
          $CMD,$ERR := exec.LookPath("...");
          ...
      - pattern-not-inside: |
          $CMD = "...";
          ...
    - patterns:
      - pattern-either:
        - pattern: |
            exec.Cmd {...,Args: []string{"=~/(sh|bash|ksh|csh|tcsh|zsh)/","-c",$EXE,...},...}
        - patterns:
          - pattern: |
              exec.Cmd {...,Args: []string{$CMD,"-c",$EXE,...},...}
          - pattern-inside: |
              $CMD,$ERR := exec.LookPath("=~/(sh|bash|ksh|csh|tcsh|zsh)/");
              ...
      - pattern-not: |
          exec.Cmd {...,Args: []string{"...","...","...",...},...}
      - pattern-not-inside: |
          $EXE = "...";
          ...
  - pattern-inside: |
      import "os/exec"
      ...
  message: >-
    Detected non-static command inside exec.Cmd. Audit the input to 'exec.Cmd'.
    If unverified user data can reach this call site, this is a code injection
    vulnerability. A malicious actor can inject a malicious script to execute
    arbitrary code.
  metadata:
    cwe:
    - "CWE-94: Improper Control of Generation of Code ('Code Injection')"
    owasp:
    - A03:2021 - Injection
    category: security
    technology:
    - go
    confidence: MEDIUM
    references:
    - https://owasp.org/Top10/A03_2021-Injection
    cwe2022-top25: true
    subcategory:
    - audit
    likelihood: LOW
    impact: HIGH
  severity: ERROR
  languages: [go]