ClawHub
Back to skill

Security audit

ARTA: Agentic Real-Time Awareness

Security checks for vulnerabilities and agentic risk

Overview

The skill is local and non-executing, but its template can reveal other sessions’ channel IDs and task descriptions without clear authorization guidance.

Install only if you are comfortable with the agent tracking and potentially reporting other active session metadata, including channel identifiers, sender names, and task summaries. Use it in environments where all users are authorized to see that cross-session activity, or add checks to restrict and redact responses before deployment.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
79% confidence
Finding

The documentation presents the implementation as purely in-memory and free of external services, but elsewhere in the same file it instructs users to set up channel detection for Telegram/Discord/Slack and optionally install an OpenClaw hook for automatic capture. That creates an intent-level contradiction in the documentation about whether the skill is purely local/simple or involves external integrations and hooks.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The template explicitly instructs the agent to disclose other active channels and tasks from separate sessions to the current user. This creates a cross-session privacy leak because channel identifiers and task descriptions may contain sensitive or identifying information, and there is no access control, consent check, or warning about privacy implications.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.