ClawHub
Back to skill

Security audit

Api Magic Gen Free

Security checks across malware telemetry and agentic risk

Overview

This is a coherent magic-api guide, but users should treat it as a tool for creating live database-backed endpoints and apply normal production controls.

Install this only if you intentionally work with magic-api. Use staging first, protect or disable the Web UI in production, review scripts before saving because endpoints go live immediately, restrict database permissions, use parameterized SQL and hashed passwords, and avoid exposing arbitrary Java classes or Spring beans unless explicitly reviewed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (4)

Intent-Code Divergence

Medium
Confidence
91% confidence
Finding
The skill claims the free edition does not support certain advanced extensibility, yet later states scripts can import Java classes and invoke Spring beans. In a script-to-live HTTP framework, that inconsistency can mislead users into exposing arbitrary application internals or privileged beans through dynamically editable endpoints, substantially increasing remote code execution and sensitive-function exposure risk.

Vague Triggers

Medium
Confidence
80% confidence
Finding
The activation language is broad ('API integration', 'system connection', 'Webhook configuration') and does not define clear boundaries for when the skill should engage. In an agent environment with read/write/exec tools, overly broad triggering can cause the skill to activate for unrelated tasks and steer the agent toward risky scripting, configuration, or execution workflows without sufficient user intent verification.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The capability coverage uses a very broad keyword list containing generic terms like 'api', 'Use', 'when', 'Webhook', and 'system connection'. Such loose matching raises the chance of accidental invocation in unrelated contexts, which is dangerous because the skill advertises executable and write-capable workflows that may alter systems or produce live endpoints.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The documentation highlights that saving a script makes the HTTP interface immediately live, but the warning about production exposure appears much later and is easy to miss. In this context, immediate publication of dynamically scripted endpoints can expose unauthenticated APIs, unsafe database operations, or broken auth logic instantly, making accidental deployment mistakes high impact.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.