ClawHub
Back to skill

Security audit

openspace-llm-xiaowei

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly does what it says, but its README publishes a full-looking MiniMax API key and gives weak warnings about sending prompts to an external LLM service.

Review before installing. Do not use the published README key; treat it as exposed and unsafe. Configure your own MiniMax credential securely, verify the `openspace` dependency and proxy settings, and avoid sending secrets, private files, regulated data, or proprietary text unless you trust the external provider and account configuration.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Context-Inappropriate Capability

High
Confidence
99% confidence
Finding
The README contains what appears to be a live MiniMax API key and explicitly instructs users to place it in environment variables. Exposing real credentials in documentation enables immediate unauthorized use, quota theft, billing abuse, and possible compromise of any data sent through that account. In the context of an LLM integration skill, there is no legitimate need to publish a concrete secret.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The README directs users to send prompts and text to an external API using environment-based credentials but does not warn that user inputs may leave the local environment and be processed by a third-party service. This creates a meaningful privacy and compliance risk because users may unknowingly submit sensitive data, especially in an agent-skill context where prompts may include workspace or user content.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill encourages users to send prompts to the MiniMax external API and configure an API key, but it does not clearly warn that prompts, potentially including sensitive user data, will be transmitted to a third-party service. In an agent setting, users may assume local processing unless disclosure is explicit, creating privacy and data-handling risk.

Static analysis

No suspicious patterns detected.