ClawHub
Back to skill

Security audit

Claude终端复用工具

Security checks for vulnerabilities and agentic risk

Overview

This tmux helper is not clearly malicious, but it asks for shell/file authority and API-key use while its scope and disclosures are inconsistent.

Review before installing. Use this only for explicit tmux session/window tasks, avoid granting it broad project-management activation, and do not provide API keys unless the publisher explains exactly what service is called and why. Confirm before allowing it to close sessions, paste into terminals, write files, or run arbitrary shell commands.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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
Findings (4)

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The skill repeatedly claims to be 'instruction-only' and to do only what it advertises, yet the document also declares exec/write tools, API-key configuration, API calls, file processing, and command execution. This mismatch can mislead users and downstream agents into granting broader trust or permissions than intended, increasing the chance of unsafe command execution or data exposure.

Intent-Code Divergence

Medium
Confidence
88% confidence
Finding
The document gives contradictory statements about synchronization support, saying in one place that tmux synchronization is unsupported while elsewhere claiming synchronization features for collaboration. In security-sensitive tooling, contradictory capability claims can cause operators or agents to rely on nonexistent safeguards or enable workflows under false assumptions.

Vague Triggers

High
Confidence
92% confidence
Finding
The description activates on broad scenarios such as project management, task planning, progress tracking, and team collaboration, which are far outside a tmux helper's narrow scope. Overbroad invocation criteria can cause the skill to be selected in inappropriate contexts, where its exec/write capabilities may be exposed to unrelated user data or prompts and increase the attack surface.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill exposes read, exec, and write capabilities but does not give a clear, prominent warning that it can run shell commands and modify files. Users or orchestrating agents may invoke it under the mistaken belief that it is a passive helper, which materially increases the risk of unintended system changes, destructive commands, or leakage through generated files/logs.

Static analysis

No suspicious patterns detected.