Back to skill

Security audit

agentshield.live verify

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because its validation workflow may send user-provided information to a third party without warning or authorization.

Before installing, consider whether you are comfortable with validation inputs being sent to agentshield.live. Avoid submitting sensitive financial information unless you understand what will be shared and have authorized it.

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
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)

The verification workflow sends user-provided validation inputs to a third-party service without a privacy warning.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The workflow directs the agent to submit validation inputs to agentshield.live without requiring disclosure or user approval for sharing potentially sensitive banking, invoice, or payment information. Although remote validation is the skill's stated purpose, its broad 'before you act' triggers could cause disclosure during tasks where the user has not authorized this third-party processing; HTTPS and payment-signing safeguards do not address that privacy risk.

Content

No source excerpt is available for this finding.