ClawHub
Back to skill

Security audit

admapix-free

Security checks for vulnerabilities and agentic risk

Overview

This is mostly a disclosed AdMapix API lookup skill, but it asks for file-write and broad command capability beyond its stated read-only API purpose without clear limits.

Review before installing. Use it only if you are comfortable sending AdMapix queries and your ADMAPIX_API_KEY to the AdMapix API. The main risk is not clear malware; it is overbroad declared authority, especially write access and loosely described command/file capabilities that are not necessary for a simple API lookup skill.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Description-Behavior Mismatch

High
Confidence
93% confidence
Finding
The skill repeatedly describes itself as a thin raw-JSON passthrough client for a remote API, but later advertises file processing, read/write, and broader automation capabilities unrelated to that scope. This scope drift is dangerous because agents or users may authorize filesystem actions they would not expect from an API lookup client, increasing the chance of unintended local data access or modification.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
A read-only API client has no clear need for write access, yet the skill declares the write tool. Unnecessary write capability violates least privilege and can enable accidental or unauthorized file creation, overwrite, or persistence on the host environment if the agent chooses to use that tool.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill markets file write capabilities without clearly limiting scope, warning about system impact, or tying them to the stated API-only purpose. In an agent context, that omission can cause users to unknowingly permit persistence or local file modification, which is especially risky when the skill otherwise presents itself as a simple data passthrough client.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The documentation advertises command execution but does not clearly warn users about potential system effects, network activity, or how execution is constrained. For agent-operated skills, undisclosed exec capability increases risk of unexpected commands, environmental inspection, or side effects beyond simple API retrieval.

External Transmission

Medium
Category
Data Exfiltration
Content
| curl 或等价 HTTP 客户端 | 命令行工具 | 必需 | 系统自带或包管理器安装 |

### 可用性分类
- **分类**: MD+EXEC（Markdown 指令驱动，需 exec 执行 curl 命令）
- **说明**: 基于自然语言指令驱动 Agent 调用 AdMapix API，透传原始结构化 JSON

## 能力梳理
Confidence
88% confidence
Finding
The skill explicitly transmits user-derived query parameters and an API key to an external third-party service. External transmission is expected for an API client, but it remains security-relevant because sensitive prompts, identifiers, or business research terms may be sent off-platform, and the broad natural-language-to-parameter mapping raises the chance of transmitting more than users realize.

Static analysis

No suspicious patterns detected.