Git高级操作免费版

• Outcome: Review
• Latest audit: Aug 29, 2026 · 4:49 AM UTC
• Version: 1.0.3

Overview

This Git helper is not clearly malicious, but it needs review because its activation scope is too broad and some Git recovery examples can erase local work.

Install only if you want an agent to help run advanced Git workflows and you are comfortable reviewing commands before execution. Treat any rebase, cherry-pick, worktree removal, or reset --hard command as potentially disruptive; check git status first, preserve uncommitted work, and avoid using this skill for non-Git AI or agent-orchestration tasks.

SkillSpector

Vulnerability Patterns

• Tool Misuse: Tool Parameter Abuse, Chaining Abuse, Unsafe Defaults
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (6)

Description-Behavior Mismatch

Medium

• Confidence: 96% confidence
• Finding: The manifest and description materially blur the scope of the skill: it is presented as a Git advanced-operations helper, but also marketed for generic AI/Agent/LLM orchestration use. This can cause the agent to invoke a command-execution-capable skill outside its intended domain, increasing the chance of unsafe or irrelevant exec actions under ambiguous prompts.

Vague Triggers

High

• Confidence: 97% confidence
• Finding: An overly broad trigger that says to use the skill for general AI-model, agent-orchestration, or LLM application scenarios can cause inappropriate activation of a skill that has exec capability. In context, this increases the risk that non-Git user requests get funneled into shell-capable behavior without clear domain restrictions.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The trigger section lacks precise invocation boundaries, so the agent may choose this skill whenever an LLM or agent workflow is mentioned, even if the task is unrelated to Git. Because the skill allows exec, weak invocation constraints materially raise the chance of unintended command execution in the wrong context.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill documents destructive commands such as 'git reset --hard' in a recovery workflow without a prominent warning that uncommitted local changes can be permanently lost. In an exec-enabled agent context, users may copy or authorize these commands without understanding the irreversible consequences, leading to data loss.

Tool Parameter Abuse

High

• Category: Tool Misuse

Content

git reflog --date=relative

# 恢复到指定状态
git reset --hard HEAD@{2}      # 回到 2 步之前的状态
git reset --hard ghi789        # 回到指定提交

# 恢复误删的分支

• Confidence: 98% confidence
• Finding: The documented use of 'git reset --hard HEAD@{2}' is a destructive parameter choice that can discard local modifications and move repository state abruptly. In a skill that exposes exec and presents command templates for direct use, this is dangerous because users or agents may execute it as a routine recovery step without backup or confirmation.

Tool Parameter Abuse

High

• Category: Tool Misuse

Content

# 恢复到指定状态
git reset --hard HEAD@{2}      # 回到 2 步之前的状态
git reset --hard ghi789        # 回到指定提交

# 恢复误删的分支
git branch recovered-branch abc123

• Confidence: 98% confidence
• Finding: The example 'git reset --hard ghi789' normalizes a high-risk destructive command that can permanently remove local work if used incorrectly. Given the skill's exec capability and broad activation language elsewhere, this increases the likelihood of unsafe repository modification under ambiguous or inexperienced use.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.