ClawHub
Back to skill

Security audit

four-word-phrases

Security checks for vulnerabilities and agentic risk

Overview

This is a simple English phrase-learning skill with no code execution, persistence, credential access, or hidden data movement.

Installers should expect a Chinese-language phrase helper that may activate on broad English-learning requests. Review the phrase quality and trigger wording if precise activation is important, but the inspected artifacts do not show unsafe permissions or hidden behavior.

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 (4)

Vague Triggers

Medium
Confidence
96% confidence
Finding
The skill advertises activation on very generic terms such as '日常口语', '常用短句', and '实用英语', which can appear in many normal conversations and cause the skill to trigger unexpectedly. While this is not directly code-execution or data-exfiltration behavior, over-broad activation increases the chance of unintended tool use and unnecessary file reads in unrelated contexts.

Vague Triggers

Medium
Confidence
97% confidence
Finding
The activation condition block uses ambiguous, high-frequency keywords without clear constraints or prioritization, making accidental invocation likely. In agent systems, this can create prompt-routing instability and may cause the assistant to read reference files or follow skill instructions when the user did not intend to use this skill.

Skill Enumeration

Medium
Category
Agent Snooping
Content
| 文件 | 路径（直接用于 read_file） | 说明 |
|------|--------------------------|------|
| SKILL.md | `skills/four-word-phrases/SKILL.md` | 本文件（已加载） |
| phrases.md | `skills/four-word-phrases/refs/phrases.md` | **完整短句库（必读）** |

**重要**：收到用户相关请求后，第一步务必调用：
Confidence
80% confidence
Finding
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Natural-Language Policy Violations

Low
Confidence
88% confidence
Finding
This markdown content forces a specific language/locale for users by providing the skill material only in Chinese, while the policy requires avoiding language constraints unless there is opt-in or a justified regional scope. No alternative language option or justification is given anywhere in the file.

Static analysis

No suspicious patterns detected.