ClawHub
Back to skill

Security audit

Simla.com

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Simla.com connector skill with some imprecise documentation, but no evidence of hidden, destructive, or unrelated behavior.

Install only if you want an agent to work with your Simla.com customer and order data through OOMOL. Review prompts carefully before allowing create or edit actions, and treat the documented oo CLI setup and OOMOL account connection as part of the trust boundary.

SkillSpector

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 (2)

Intent-Code Divergence

Medium
Confidence
91% confidence
Finding
The action list marks `get_order` and `get_order_statuses` as `[write]` even though their names and descriptions indicate read-only retrieval operations. This mismatch can cause an agent to apply the wrong safety policy, either needlessly blocking legitimate reads or, more importantly, normalizing inaccurate safety labels so future truly mutating actions may be misclassified and executed under the wrong assumptions.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger phrase says to use this skill for "ANY Simla.com request," which is overly broad and can cause the agent to route all Simla-related tasks through a high-privilege integration without considering narrower, safer alternatives. In a skill that supports both reads and writes, broad invocation language increases the chance of unintended use, overreach, and accidental state-changing operations.

Static analysis

No suspicious patterns detected.