ClawHub
Back to skill

Security audit

aws-wechat-article-writing-free

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed WeChat article-drafting workflow that uses local article files and a user-configured LLM endpoint, with no artifact-backed evidence of hidden or destructive behavior.

Install only if you are comfortable with article inputs and configuration being sent to the LLM endpoint you configure, and use a dedicated API key. Check which write.py will run in your environment because this package does not include that script.

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 (1)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The invocation/workflow instructions are broad enough that an agent may auto-initiate multi-step actions such as reading config files, writing configuration back to disk, creating directories, invoking `write.py`, and sending article content plus configuration to an external LLM endpoint without a narrowly defined user trigger boundary. In agent environments with `read`, `write`, and `exec` tools, this increases the risk of unintended side effects and data exfiltration because ordinary user requests about writing could be interpreted as authorization to modify files and call remote services.

Static analysis

No suspicious patterns detected.