Back to skill

Security audit

OpenClaw Git Workflow

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local Git helper that reads repository diffs and branch metadata, with documentation that overstates some unimplemented mutating features but no evidence of hidden, destructive, persistent, or exfiltrating behavior.

Before installing, treat this as a local Git inspection helper and review diffs before sharing generated output because repository diffs can contain sensitive code or secrets. Do not rely on the advertised smart-commit or branch-delete workflows unless the package is updated to implement them with clear previews and confirmation.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (12)

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The documented behavior claims broad Git automation, including PR generation and branch management, while the analysis indicates the implementation does not match and may still access repository contents without declared permissions. This mismatch is dangerous because users may trust the skill to perform only documented actions, reducing scrutiny over hidden or undeclared access to code and metadata.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding

The skill documentation clearly instructs users to run shell-based commands and describes automation that can stage, commit, and delete branches, but it declares no tool scope or permissions. This creates an authorization/transparency gap: users and hosting systems are not warned that repository state and local shell access may be used.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill advertises automated staging, committing, and branch deletion without an upfront safety warning, which can lead users to trigger destructive repository changes unintentionally. In a Git workflow context, these actions can alter history, include unintended files in commits, or delete branches that users expected to keep.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/commit_generator.py (reported line 67)May include surrounding context.

python
def _get_staged_diff(self) -> str:
        """Get staged diff"""
        result = subprocess.run(
            ['git', 'diff', '--cached'],
            capture_output=True,
            text=True,

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/commit_generator.py (reported line 77)May include surrounding context.

python
def _get_changed_files(self) -> List[Dict[str, str]]:
        """Get list of changed files"""
        result = subprocess.run(
            ['git', 'diff', '--cached', '--name-status'],
            capture_output=True,
            text=True,

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/main.py (reported line 37)May include surrounding context.

python
def run_pr_description(args):
    """Generate PR description"""
    # Get branch diff summary
    result = subprocess.run(
        ['git', 'log', f'{args.base}..{args.head}', '--oneline'],
        capture_output=True,
        text=True,

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/main.py (reported line 47)May include surrounding context.

python
commits = result.stdout.strip().split('\n') if result.stdout else []

    # Get changed files
    result = subprocess.run(
        ['git', 'diff', f'{args.base}...{args.head}', '--name-status'],
        capture_output=True,
        text=True,

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/main.py (reported line 93)May include surrounding context.

python
def run_branch_strategy(args):
    """Show branch strategy suggestions"""
    # Get current branch
    result = subprocess.run(
        ['git', 'branch', '--show-current'],
        capture_output=True,
        text=True,

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/main.py (reported line 102)May include surrounding context.

python
current_branch = result.stdout.strip()

    # Check if behind main
    result = subprocess.run(
        ['git', 'rev-list', '--count', f'HEAD..origin/{args.base}'],
        capture_output=True,
        text=True,

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/main.py (reported line 135)May include surrounding context.

python
def run_check_branches(args):
    """Check and clean branches"""
    # Get all branches
    result = subprocess.run(
        ['git', 'branch', '-a'],
        capture_output=True,
        text=True,

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/main.py (reported line 145)May include surrounding context.

python
branches = [b.strip().strip('* ') for b in result.stdout.strip().split('\n') if b.strip()]

    # Get default branch
    result = subprocess.run(
        ['git', 'symbolic-ref', 'refs/remotes/origin/HEAD'],
        capture_output=True,
        text=True,

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
96% confidence
Finding

The file presents core instructional content in Chinese, including the summary and feature descriptions, without indicating that this language is optional or region-specific. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.