ClawHub
Back to skill

Security audit

NebulaMind Agent Council

Security checks for vulnerabilities and agentic risk

Overview

This skill openly uses a NebulaMind API key to register an agent and submit astronomy jury votes; that behavior matches its stated purpose but users should treat the key and voting actions carefully.

Install only if you want an agent to participate in NebulaMind jury voting. Store the API key like a password, run the script with --dry-run first, keep NEBULAMIND_API unset unless you intentionally trust another endpoint, and remember that submitted votes and reputation changes are external account actions.

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
95% confidence
Finding
The skill documents use of environment variables and outbound network access but does not declare those capabilities. Hidden or undeclared capabilities reduce user and platform visibility into what the skill can access and do, especially because it handles an API key and sends data to an external service. In a skill that registers agents and votes publicly, this matters because secrets and user-controlled metadata may be transmitted off-platform.

Intent-Code Divergence

Medium
Confidence
83% confidence
Finding
The document claims a minimal scope, then references an additional challenge-submission capability that is not otherwise specified. Undeclared functionality is risky because it can expand what the skill may cause a user or agent to do, including submitting public challenges to a third-party service without clear parameters, safeguards, or permission disclosure. This is especially relevant in an agent skill because users may rely on the stated narrow scope when authorizing use.

Static analysis

No suspicious patterns detected.