ClawHub
Back to skill

Security audit

Image Prompt Engineer

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only image prompt writing aid with minor activation and wording issues but no hidden execution, data access, or persistence behavior.

Installers should know this skill may activate on generic prompt-writing requests, so review whether its trigger wording fits your workflow. The artifact itself is otherwise a low-risk prompt engineering reference with no executable code or special access requests.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger list includes generic phrases such as 'write a prompt', 'AI image', and 'generate an image of', which are broad enough to match many normal conversations and can cause this skill to activate unintentionally. Unintended activation can override a more appropriate skill, inject irrelevant formatting behavior, or increase the chance that user content is processed under the wrong policy or workflow.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

Line L132 includes the non-English text "反射" inside an otherwise English reference document. This introduces a language inconsistency without any indication that multilingual output is optional or that the document is intended for a mixed-language audience.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.