Back to skill

Security audit

Slop Cop

Security checks for vulnerabilities and agentic risk

Overview

This skill is a visual QA helper for images and design assets, and its file access and vision-tool use fit that purpose.

Install this if you want an agent to critique images, screenshots, and design variants before shipping. Only provide image files, URLs, or screenshots you are comfortable having inspected by the agent and vision tool.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (4)

Hidden Instructions

High
Category
Prompt Injection
Confidence
60% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · README.md (reported line 117)May include surrounding context.

md
---

<p align="center">
  <i>Built by <a href="https://github.com/chchchadzilla">@chchchadzilla</a> with crow-energy assist from José 🐦‍⬛</i>
</p>

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
55% confidence
Finding

Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Content

Scanner excerpt · references/vision-prompt-template.md (reported line 58)May include surrounding context.

md
Be terse. Do not apologize. Do not add a preamble.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The README tells users the AI will 'load the skill automatically' and lists broad natural-language prompts that can trigger it, but it does not define clear activation boundaries, required user confirmation, or scope constraints. In an agentic environment, vague auto-invocation language can cause the skill to run on unintended files or contexts, increasing the chance of unauthorized analysis, surprise tool use, or overbroad access to local assets.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The invocation guidance includes broad phrases like general requests for a 'second opinion' or whether something is 'safe to ship,' which can overlap with ordinary design, review, or marketing tasks outside this skill’s narrow purpose. In an agent environment, that increases the chance of over-invocation, causing the skill to intercept unrelated requests and influence decisions where visual-forensics rigor may not be appropriate.

Content

No source excerpt is available for this finding.