ClawHub
Back to skill

Security audit

Workspace Git Setup

Security checks for vulnerabilities and agentic risk

Overview

This skill transparently sets up local Git tracking and can create a local commit, with no hidden upload, installer, or privilege escalation found.

Install only if you want a tool that may initialize Git and locally commit the selected workspace. Run it with --dry-run or --audit first, inspect .gitignore and git status, and remember that a local commit can still include files not covered by the ignore rules even though the skill does not push to a remote.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

High
Confidence
98% confidence
Finding
The trigger definition is excessively broad and includes both generic phrases like 'back up my workspace' and a catch-all clause for 'any other phrasing' related to version tracking or repo health. This can cause the skill to activate in unintended contexts, leading to unrequested repository initialization, .gitignore overwrites, config changes, or commits in arbitrary directories, which is a real security and safety risk even though the skill's stated purpose is benign.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The script performs `git add -A` followed by an automatic commit of all staged changes in the target workspace without a mandatory explicit confirmation immediately before staging/committing. In a tool whose purpose is 'one-command Git setup', this can accidentally capture source files, generated artifacts, or sensitive files not covered by the default `.gitignore`, especially because the script also supports arbitrary workspace paths and only partially audits for secrets.

Static analysis

No suspicious patterns detected.