ClawHub
Back to skill

Security audit

local-skill

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small local release-check helper with disclosed scripts and no evidence of hidden access, persistence, or harmful behavior.

Before installing, treat this as a local release workflow helper: review the package metadata and scripts, and only run the publish script from the repository root after confirming the target package/version. No hidden or high-impact behavior was found in the inspected artifact.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The declared description says this skill is a local repository for publish and pre-publish inspection and specifically provides inspectSkill and summarizeSkill utilities. The supplied code does not implement those utilities or repository inspection logic. Instead, it is a publish helper script that invokes another check script and reads package metadata to announce publishing. While pre-publish checking loosely relates to the description, the primary behavior shown is package publishing workflow support, not skill inspection utilities. This is a material description-to-behavior mismatch.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The description claims the skill provides inspection and summarization utilities for local skill repository publish workflows. However, the supplied code chunk does not implement any such functionality; it only exports a name and version object. There are no undeclared harmful capabilities, resource accesses, or triggers, but the actual behavior in this chunk is materially different from the declared primary purpose because the promised utilities are absent.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.