ClawHub
Back to skill

Security audit

Ai Assistant

Security checks for vulnerabilities and agentic risk

Overview

This is mostly a document-analysis skill, but it asks for broad write, command, API, and credential-related capabilities without clear limits or user-control rules.

Review before installing. Use this only if you are comfortable granting a document-analysis skill read/write and command-execution capability, and avoid giving it sensitive contracts, credentials, or private business documents unless you can enforce explicit user approval for API calls, shell commands, and file writes.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The skill is presented as a document-analysis tool, but later expands into file handling, API integration, and command execution. That capability drift is dangerous because users may invoke the skill expecting passive analysis while the agent is implicitly encouraged to perform state-changing or externally connected actions, increasing risk of unintended command execution, data disclosure, or file modification.

Vague Triggers

High
Confidence
95% confidence
Finding
The activation language is extremely broad ('use when AI model calling, intelligent dialogue, agent orchestration, LLM applications are needed'), which can cause the skill to trigger in many unrelated contexts. In a skill that declares read/exec/write tools, overbroad activation materially increases the chance that sensitive documents or system actions are routed through this skill without a narrowly justified reason.

Missing User Warnings

High
Confidence
96% confidence
Finding
The skill advertises file writing, external API usage, and command execution without prominent warnings about data exposure, system modification, or user-consent requirements. In an agent setting, these are high-risk capabilities because they can transform a benign document-review workflow into one that alters files, sends sensitive content to third parties, or runs shell commands based on untrusted input.

Static analysis

No suspicious patterns detected.