ClawHub
Back to skill

Security audit

aicut-fake-text

Security checks for vulnerabilities and agentic risk

Overview

The skill creates chat videos through a disclosed external service, with clear approval steps for spending and no evidence of hidden access.

Use fictional names and avoid sensitive conversations: submitted text is processed externally. Review the script and total price before approving generation. The CLI retains a local sign-in session until you log out or revoke access.

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)

The generation workflow lacks a privacy warning that chat text and contact information are submitted to external services.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The workflow instructs the agent to submit the messages and contact name to aicut for video generation, with OpenAI or ElevenLabs selected for speech generation. It requires approval for token spending but does not warn that supplied text will be processed externally, which matters if the user includes real conversations or personal information.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.