ClawHub
Back to skill

Security audit

meta-game-negotiation

Security checks for vulnerabilities and agentic risk

Overview

The skill is a low-authority negotiation helper with some overstated self-improvement claims, but no evidence of hidden access, exfiltration, destructive behavior, or unsafe automatic execution.

Install only if you are comfortable with a skill that is more of a generated negotiation prompt plus a small local stats writer than a fully implemented self-verifying negotiation system. Do not rely on its claimed self-verification, orchestration, or continuous evolution without independently checking outputs.

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
77% confidence
Finding
The skill advertises behavior that includes writing to persistent state (`self-reflection-loop`, learner injection, statistics/memory updates) but declares no permissions. Undeclared write capability is dangerous because it obscures the skill’s real side effects, preventing users or a runtime policy engine from making an informed trust decision and enabling unauthorized persistence or tampering if the skill is executed.

Tp4

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding
The skill description makes strong claims about advanced negotiation, self-verification, orchestration, and continuous self-evolution that are not supported by the actual behavior. This mismatch is dangerous because operators may rely on nonexistent safety checks or decision logic, causing unsafe automation, misplaced trust, and unintended execution of a purportedly higher-assurance skill.

Static analysis

No suspicious patterns detected.