ClawHub
Back to skill

Security audit

Automator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed workflow-automation guide whose recurring tasks and external notifications fit its stated purpose, with user caution needed for schedules and outbound sharing.

Before installing or using this skill, treat scheduled workflows as persistent automations: test with non-production agents first, review recipients and channels, use least-privilege credentials, add approval gates for sensitive reports or public posting, and monitor or pause workflows when no longer needed.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The skill explicitly encourages scheduled automations that can send emails, post to external platforms, and perform backups, but it does not warn users that these actions may transmit data off-system or create recurring side effects without continuous review. In an automation skill, omission of consent, review, or destination-scope guidance increases the risk of unintended data disclosure or disruptive automated actions.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
80% confidence
Finding

The skill promotes recurring scheduled tasks and cross-platform automation, which inherently persist beyond the current session and can continue operating without immediate user supervision. That persistence becomes risky when combined with actions like social posting, monitoring, or backups, because a misconfigured workflow may repeatedly leak data or perform unintended actions over time.

Content

Scanner excerpt · SKILL.md (reported line 24)May include surrounding context.

md
- "Automate my daily report generation"
- "Create a workflow that monitors prices and alerts me"
- "Set up a multi-step data processing pipeline"
- "I need to schedule recurring tasks with dependencies"
- "Automate my social media posting across platforms"
- "Create an approval workflow for my team"
- "Set up automated backups with notifications"

Session Persistence

Medium
Category
Rogue Agent
Confidence
72% confidence
Finding

This example shows an automated analytics pipeline that culminates in distribution over email and Slack, implying repeated outbound sharing of generated content. Without nearby warnings about review, recipient validation, and sensitivity of report contents, persistent distribution workflows can repeatedly expose internal data to the wrong audience or external services.

Content

Scanner excerpt · SKILL.md (reported line 187)May include surrounding context.

md
agent: "analyst"
    depends_on: [transform]

  - report: "Create PDF report"
    agent: "reporter"
    depends_on: [analyze]

Static analysis

No suspicious patterns detected.