Back to skill

Security audit

GitHub开发规范工具

Security checks for vulnerabilities and agentic risk

Overview

The skill is a broad GitHub development helper with project read, write, and shell execution authority, but its safety limits are not clearly scoped or enforceable.

Review this skill before installing if you expect narrow GitHub workflow help. It may be useful for development automation, but only use it in repositories where file edits and shell commands are acceptable, and require explicit confirmation for command execution or project-wide changes.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The activation guidance is very broad, covering common coding, debugging, testing, deployment, project management, and development-assistance scenarios. Because the skill also requests powerful tools including exec and write, this broad trigger surface increases the chance the skill is invoked in unrelated or insufficiently scoped contexts, which can expose repositories, credentials, or system state to unnecessary actions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The description states '支持中文交互,无需复杂配置即开即用', presenting Chinese as the default interaction mode without offering a language choice. This is a natural-language locale policy issue because the skill appears to force or assume a specific language rather than letting the user opt in.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill advertises that only whitelist commands will be executed, but the file provides no actual whitelist, validation rules, or enforcement logic. In a skill that exposes the exec tool, this can create a false sense of safety and lead operators or downstream agents to execute arbitrary shell commands under the assumption that restrictions exist.

Content

No source excerpt is available for this finding.