ClawHub
Back to skill

Security audit

git-batch-commit

Security checks across malware telemetry and agentic risk

Overview

This Git commit helper is coherent and disclosed, but users should treat it as a repository-mutating tool and review prompts before publishing or pushing anything.

Install only if you want an agent to reorganize staged Git changes into commits. Use dry-run or the default confirmation flow for normal work, and be cautious with --yes or with repositories you do not trust. Read any prompt about ClawHub/SkillHub publishing or subtree pushing carefully because those actions can upload or push repository content.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill declares that it invokes external commands via subprocess, including git and potentially clawhub/skillhub after user confirmation, but the metadata does not declare corresponding shell or command-execution permissions. This creates a capability/permission mismatch that can mislead users and security controls about what the skill is able to do, especially because it can modify repository state and potentially publish content to external platforms.

Unvalidated Output Injection

High
Category
Output Handling
Content
This analyzes the git diff content.
    """
    try:
        result = subprocess.run(
            ['git', 'diff', '--cached', filepath],
            capture_output=True,
            text=True,
Confidence
69% confidence
Finding
subprocess.run( ['git', 'diff', '--cached', filepath], capture_output

Unvalidated Output Injection

High
Category
Output Handling
Content
if not files:
        return True
    try:
        subprocess.run(
            ['git', 'add'] + files,
            capture_output=True,
            check=True
Confidence
84% confidence
Finding
subprocess.run( ['git', 'add'] + files, capture_output

Unvalidated Output Injection

High
Category
Output Handling
Content
if not files:
        return True
    try:
        subprocess.run(
            ['git', 'reset', 'HEAD'] + files,
            capture_output=True,
            check=True
Confidence
86% confidence
Finding
subprocess.run( ['git', 'reset', 'HEAD'] + files, capture_output

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.