ClawHub
Back to skill

Security audit

Daily English Card

Security checks for vulnerabilities and agentic risk

Overview

The skill has a coherent English-learning purpose, but it automates recurring WeChat delivery and IMA archival through fixed account details and local credential helpers without enough user-specific scoping or controls.

Review before installing. Replace the hard-coded WeChat target/account and IMA IDs with your own, inspect the referenced local Python and Node scripts because they are not bundled here, confirm what gets uploaded and retained, and only enable the daily cron job after you know how to pause or remove it.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The README documents automatic delivery of generated images to a specific WeChat recipient and automatic upload to Tencent IMA, but provides no consent flow, privacy notice, or data handling warning. Even if the content is educational, this creates undisclosed external data transmission and persistent archival behavior that could surprise users or leak generated material and metadata to third-party services.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The cron payload explicitly instructs the agent to send the generated image via the message tool to a fixed WeChat target, with requirements that effectively force external transmission. This is dangerous because it automates outbound messaging without an approval checkpoint, increasing the risk of unwanted delivery, privacy issues, or misuse if the generated content or destination is wrong.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger list contains very broad everyday phrases such as “口语”, “学英语”, and “背单词”, which can cause the skill to activate unintentionally during normal conversation. Because the skill performs outbound actions automatically, accidental activation could lead to unsolicited WeChat messages or external data handling without clear user intent.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly describes automatic WeChat delivery and IMA archival, but it does not present a clear user-facing warning or consent boundary around external transmission. Even if the payload is usually benign learning content, automatic outbound delivery to fixed third-party destinations creates a privacy and misuse risk if triggered unexpectedly or configured incorrectly.

Static analysis

No suspicious patterns detected.