ClawHub
Back to skill

Security audit

using-overpowered

Security checks for vulnerabilities and agentic risk

Overview

This is a routing-only skill that tells the agent how to choose among related workflow skills, with no bundled code or hidden persistence.

Before installing, understand that this skill can make the agent route complex work through several companion Overpowered skills, including automation or capability-generation paths when justified. Review those companion skills separately, especially for side-effecting automation, but this artifact itself is disclosed, non-executable, and includes controls for risky work.

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
  • 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
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill description is extremely broad and positions this skill as an entry point for many kinds of knowledge-work and automation tasks. That overbroad routing scope can cause the orchestrator to activate the skill in contexts beyond its safe or intended boundaries, increasing the chance of unnecessary tool use, inappropriate delegation to other powerful skills, or bypass of narrower, safer workflows.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 98)May include surrounding context.

md
## Interaction rule

If a skill can discover a fact from tools or artifacts, do not ask the user for that fact. Ask humans for intent, decisions, unavailable tacit knowledge, authority, or approval required by risk policy. Ask the minimum decision-changing question, and do not require perfect specification before reversible work.

## Adaptive capability rule

Static analysis

No suspicious patterns detected.