ClawHub
Back to skill

Security audit

kaios

Security checks for vulnerabilities and agentic risk

Overview

The skill is coherent for creating OpenClaw agents and Discord channels, but it gives broad authority to create persistent agents, memory, cron jobs, gateway changes, and Discord mutations without strong scoping or retention safeguards.

Review this carefully before installing. It is not evidence of malware, but it can create autonomous agents, modify OpenClaw gateway configuration, manage Discord channels, write memory files, and set up scheduled behavior. Use it only in a controlled workspace, grant Discord permissions narrowly, avoid broad workspace paths, and confirm any cron jobs or kept agent sessions are actually wanted.

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 (2)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill description is extremely broad, covering identity configuration, persistent memory, task scheduling, multi-agent orchestration, and workspace management without clear activation boundaries or user-consent gates. In an agent environment, vague triggers can cause the skill to activate in unrelated contexts and introduce stateful behavior, autonomous actions, or data handling beyond what the user intended.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs the agent to maintain persistent memory files and to automatically check and execute recurring tasks, but it provides no user-facing warning, consent model, retention policy, or limits on external actions. In this context, the combination of long-term memory and autonomous task execution is especially risky because the skill is explicitly designed to give agents durable state and proactive behavior, which can lead to privacy leakage, unintended data retention, and unauthorized actions.

Static analysis

No suspicious patterns detected.