ClawHub
Back to skill

Security audit

pony-image

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent image-generation skill that sends prompts and images to a disclosed remote backend, which is expected for its purpose but worth using carefully with sensitive content.

Install only if you are comfortable sending image prompts, product descriptions, reference image URLs, and possible Base64 image data to the listed Supabase service for processing. Avoid using confidential, regulated, or unreleased business assets unless the provider's retention and privacy terms are acceptable to you.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill instructs the agent to send user prompts, product descriptions, and potentially user-supplied image data to a remote Supabase endpoint, but it does not clearly disclose this external transmission to the user or require consent. In an image-generation skill, these inputs can contain sensitive business assets, proprietary product photos, or personal data, so the omission creates a real privacy and data-handling risk.

External Transmission

Medium
Category
Data Exfiltration
Content
**执行方式**：收集完信息后，执行以下命令：

```bash
curl -s -X POST "$BASE_URL/generate-image" \
  -H "Authorization: Bearer $PONY_SUPABASE_ANON_KEY" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
96% confidence
Finding
This command performs an external POST request to a remote backend with user-provided prompt content and authorization headers, confirming actual data exfiltration from the local agent context to an external service. In this skill's context that behavior is expected for functionality, but it is still security-relevant because sensitive user content may be transmitted off-platform without adequate notice, validation, or governance controls.

Static analysis

No suspicious patterns detected.