ClawHub
Back to skill

Security audit

ALAPI 接口助手

Security checks for vulnerabilities and agentic risk

Overview

The skill appears to use an external API token in a way that may happen from the local environment rather than only by explicit user-provided consent.

Review the skill before installing if you have ALAPI_TOKEN set locally. Install only if you are comfortable with the agent using that stored token for ALAPI requests, and prefer a version that requires explicit confirmation before any authenticated network call.

SkillSpector

Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill instructs the agent to use network access and read the `ALAPI_TOKEN` environment variable, but it declares no permissions. This creates hidden capability use that can bypass user expectations and review controls, especially because credential-backed external calls may occur without an explicit permission boundary.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The skill claims live calls occur only when the user explicitly provides a token and confirms, but later authorizes fallback to the ambient `ALAPI_TOKEN` environment variable. That mismatch can cause unintended authenticated requests and violates the stated consent model, making secret use and external data disclosure more likely in a tool that performs real network operations.

Static analysis

No suspicious patterns detected.