Back to skill

Security audit

ai-bot-access-check

Security checks for vulnerabilities and agentic risk

Overview

The scanners report a privacy concern: a user-supplied page URL is sent to an external service without a warning, so the skill is rated suspicious with medium confidence.

Before installing, consider whether you may submit private or sensitive URLs. Use only a public URL you are comfortable sharing with the external service, and note that the scan was partial because some local references could not be resolved.

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 workflow sends the user-supplied page URL to a third-party service without a privacy warning.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

The workflow sends the supplied page URL to an external service without warning about disclosure or checking for sensitive URL components. Full URLs can contain credentials, signed access tokens, private paths, or personal information that would consequently be disclosed to the service, with potential retention in request logs or caches. Using an external checker is consistent with the skill's purpose and is not evidence of malicious intent, but its public-site focus does not guarantee that every supplied URL is safe to share.

Content

No source excerpt is available for this finding.