ClawHub
Back to skill

Security audit

Autonomous Organization

Security checks for vulnerabilities and agentic risk

Overview

This skill openly aims to run autonomous agents continuously, but it gives them broad, under-scoped authority to spawn work, change memory, and organize files.

Review this before installing if you do not want an agent framework that encourages ongoing autonomous work. Use it only with tight workspace permissions, explicit task scopes, human approval for file changes or cleanup, and clear limits on memory updates and spawned sub-agents.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill explicitly defines a "Cleanup Agent" that organizes files and updates documentation, and elsewhere schedules autonomous cleanup and memory optimization tasks without any safety constraints, scope limits, confirmation requirements, or rollback guidance. In an autonomous agent framework, unattended file-maintenance behavior can cause unintended modification, deletion, or corruption of user data, making this a real operational security risk even if not overtly malicious.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill instructs users to spawn sub-agents with arbitrary tasks using sessions_spawn and states that sub-agents should execute while the orchestrator plans, but it does not warn that these agents may act unattended or require constrained permissions. In the context of an autonomous organization skill designed for continuous operation, this increases the chance of unsupervised actions, task amplification, and unsafe execution across the workspace or connected systems.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.