ClawHub
Back to skill

Security audit

switching-costs

Security checks for vulnerabilities and agentic risk

Overview

This business-strategy guide fits its stated purpose, with a minor privacy caveat when saving lessons from real use.

Review any proposed additions to the skill and save only generalized lessons without customer names, confidential pricing, or private business details.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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 (1)

Instructs the agent to persist real-use failure patterns without a privacy warning.

Medium
Category
Output Handling
Confidence
86% confidence
Finding

The instruction to add observed entries after each real use and paste the actual failure pattern encourages persisting information from user interactions into a reusable skill without consent or sanitization requirements. In this business-strategy context, those entries could contain confidential customer details, pricing, or internal decisions and later surface in shared files or unrelated sessions; however, the skill does not explicitly instruct external transmission or publication.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.