ClawHub
Back to skill

Security audit

Aws Wechat Article Writing Free

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed WeChat article-drafting helper that reads writing configuration, may call a user-configured LLM endpoint, and writes draft files, with no 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. Use a dedicated API key, verify the base_url is trusted, and expect the skill to update writing config and create draft.md files in article directories.

SkillSpector

By NVIDIA
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 (2)

Vague Triggers

Medium
Confidence
93% confidence
Finding
These lines instruct the agent to invoke the skill based on vague references to its 'related capabilities' rather than narrowly defined user intents. In an agent environment, broad activation language can cause the skill to trigger in contexts outside article writing, leading to unintended file reads, writes, or execution of `write.py` when the user did not explicitly request this workflow.

Vague Triggers

Medium
Confidence
91% confidence
Finding
This section adds highly generic capabilities like 'instruction parsing and execution', 'data processing and transformation', and 'result validation and output' that are not specific to the documented writing task. Such broad claims expand the apparent scope of the skill, increasing the chance an agent will route unrelated requests to a tool-enabled skill that can read files and run commands, which raises the risk of unintended side effects and data exposure.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.