ClawHub
Back to skill

Security audit

AFOL BrickOwl

Security checks across malware telemetry and agentic risk

Overview

This appears to be a real BrickOwl API helper, but it needs Review because one included prompt conflicts with the safer instructions and says seller inventory changes can run without approval.

Install only if you are comfortable giving the skill a BrickOwl API key that can read private store/order data and modify inventory. Before using it, treat the no-approval prompt as unsafe: require a dry run and explicit confirmation for every create, update, delete, wishlist, or bulk write.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill clearly relies on sensitive capabilities (`BRICKOWL_API_KEY` from the environment and outbound network access to the BrickOwl API) but does not declare permissions. That mismatch can bypass platform trust/consent expectations, making users or orchestration layers unaware that the skill can access credentials and transmit data externally.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The API specification explicitly requires the API key to be sent as a URL query parameter for GET requests. Query-string secrets are commonly exposed via logs, browser history, proxy/CDN telemetry, referrer leakage, and monitoring systems, so this design increases the chance of credential disclosure. In an agent skill context, this is more dangerous because the toolchain may automatically log full request URLs during execution or debugging.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The inventory update endpoint includes a delete boolean that can remove lots, but the specification provides no warning, safeguard semantics, or confirmation guidance around destructive actions. In an agent setting, ambiguous or under-guarded destructive operations raise the risk of accidental mass deletion from misprompting, parameter confusion, or automation mistakes.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The prompt explicitly states that BrickOwl write operations execute immediately with no approval flow, while exposing destructive actions such as create, update, and delete listing. In a conversational agent, this materially increases the risk of unintended marketplace changes from ambiguous prompts, prompt injection, or user misunderstanding because there is no enforced confirmation barrier before state-changing actions.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.