ClawHub
Back to skill

Security audit

Generate conventional commit messages from code changes or diff.

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only helper for drafting Conventional Commit messages, with no hidden execution, persistence, or data exfiltration behavior found.

Installers should understand that this skill helps draft commit messages and may trigger on broad Git-related phrases. If any optional Git integration is added later, require explicit user confirmation before copying to the clipboard, saving templates, reading project history, or running git commit.

A.I.G

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

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 (3)

Shadow Command Trigger

Medium
Category
Trigger Abuse
Confidence
86% confidence
Finding
The trigger phrase 'commit message' semantically overlaps with the built-in 'commit' command space, creating a shadowing risk where this skill may capture requests intended for core Git workflows. Such overlap can confuse command routing, produce unintended behavior, and reduce trust in agent determinism, particularly if users expect a native commit action rather than advisory text generation.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger list includes very generic phrases like 'commit message', 'git commit', and 'write a commit', which are likely to appear in ordinary developer conversation and tooling contexts. This can cause unintended activation of the skill, leading it to intercept unrelated requests or override more specific behaviors, especially in environments with multiple skills or built-in commands.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The activation guidance says the skill activates when the user requests a Git commit message, but it does not define exclusions, precedence, or minimum required input beyond broad examples. In a multi-skill agent, this ambiguity increases the chance of accidental invocation, misrouting user intent, or conflicting with other skills that handle Git operations more safely or more specifically.

Static analysis

No suspicious patterns detected.