ClawHub
Back to skill

Security audit

学习阶段计划

Security checks for vulnerabilities and agentic risk

Overview

This is a study-planning skill that creates a local Chinese HTML report, with minor usability and file-overwrite considerations but no evidence of malicious behavior.

Install if you want a Chinese-language study-plan report generator. Be aware it may activate on broad study-planning requests and will save the details you provide into a local HTML file that may be updated in place during revisions.

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
88% confidence
Finding
The trigger list is extremely broad and includes many common study-related phrases, so the skill may activate in situations where the user did not explicitly request file generation or structured planning. This increases the chance of unintended behavior, including unnecessary file creation and use of the skill’s constrained workflow instead of a normal conversational response.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The activation rules rely on high-level intents like asking how to review or prepare, but do not define clear non-activation boundaries. That ambiguity can cause the skill to hijack ordinary tutoring or advice requests and push users into a multi-step workflow with file output they may not expect.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill instructs the agent to write an HTML file into the workspace without any prior user notice or consent. Silent file creation can surprise users, clutter shared workspaces, and create a channel for persistence of user-derived content beyond the immediate conversation.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The iteration step authorizes direct modification of the original HTML file without warning about overwrite behavior or preserving prior versions. This can lead to accidental data loss, confusion about what changed, and destruction of earlier outputs the user may have wanted to keep.

Static analysis

No suspicious patterns detected.