ClawHub
Back to skill

Security audit

Discord Hub My

Security checks across malware telemetry and agentic risk

Overview

The skill is a small Discord webhook helper whose file access and network call are disclosed and limited to sending a user-provided message to a configured Discord webhook.

Install only if you are comfortable storing a Discord webhook URL in the skill's local .env file and sending messages through that webhook. Keep the .env file private and trusted because this implementation sources it as shell code; do not use a .env file from an untrusted source.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Credential Access

High
Category
Privilege Escalation
Content
DIR="$(cd "$(dirname "$0")" && pwd)"

# 读取 .env
if [ -f "$DIR/.env" ]; then
  set -a
  . "$DIR/.env"
  set +a
Confidence
89% confidence
Finding
Using `set -a` and sourcing `. "$DIR/.env"` exports every variable defined in the file into the process environment, and because `.` executes shell syntax, a maliciously modified .env file could run arbitrary commands. In an agent skill context, local skill files may be treated as trusted configuration, but this pattern is still dangerous if the workspace or package contents can be influenced by an attacker.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.