ClawHub
Back to skill

Security audit

Use ClawRSS

Security checks for vulnerabilities and agentic risk

Overview

This instruction-only ClawRSS skill is coherent and disclosed, but users should be deliberate before saving content, deleting feeds, marking items consumed, or sending push notifications.

Install this if you want your agent to operate ClawRSS in the current workspace. Before sensitive actions, confirm the workspace and intent, especially for deleting feeds, marking stored items consumed, saving digest content, or sending Apple push notifications whose text may appear on connected devices.

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
91% confidence
Finding
The skill enables implicit invocation without any visible trigger constraints, which can cause the agent to activate feed management, saved article access, digest handling, or push-notification functionality without sufficiently explicit user intent. In a plugin that can read persisted content and send outbound notifications, overly broad auto-invocation increases the risk of unauthorized actions, privacy leakage, and confused-deputy behavior.

Static analysis

No suspicious patterns detected.