ClawHub
Back to skill

Security audit

Tools Litiao

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple preference-tracking skill for tool choices, with disclosed persistence but no executable code or hidden data movement.

Install this only if you want the agent to remember and reuse your tool preferences across tasks. Review changes to the preference sections periodically, especially entries under Stack, Preferences, Open To, and Avoid.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Self-Modification

High
Category
Rogue Agent
Confidence
85% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · criteria.md (reported line 3)May include surrounding context.

md
# Criteria for Tool Preferences

Reference only — consult when deciding whether to update SKILL.md.

## When to Add to Stack

Unrestricted Tool Access

Medium
Category
Excessive Agency
Confidence
87% confidence
Finding

The skill explicitly authorizes use of 'ANY tool' and says user preferences should not restrict capabilities, which weakens the principle of least privilege and can enable unnecessary access to powerful or sensitive tools. In an agent setting, broad tool-use instructions can cause the model to invoke high-risk tools outside a narrowly scoped task, especially when no approval, allowlist, or risk-based constraints are defined.

Content

Scanner excerpt · SKILL.md (reported line 8)May include surrounding context.

md
## Adaptive Tool Intelligence

You can use ANY tool. Learn new ones instantly. This skill tracks user preferences, not your limits.

**Rules:**
- Default to user's known tools when they exist

Static analysis

No suspicious patterns detected.