ClawHub
Back to skill

Security audit

scaffold-mvp

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed MVP project scaffolding guide, with some install-time and repo-local agent configuration behavior users should review before running.

Install this only in a new project where you want automated MVP scaffolding. Before allowing it to proceed, review the generated .claude hooks, rules, skills, selected package versions, and git commit behavior; English-speaking users should ask for confirmations in their preferred language if the Portuguese prompts are unclear.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (1)

Natural-Language Policy Violations

Medium
Confidence
96% confidence
Finding
The skill hard-requires Portuguese prompts and confirmations even though its trigger conditions explicitly include English-speaking users. This can cause user misunderstanding or incorrect consent during stack selection, increasing the chance of unintended actions or unsafe approvals in an automation workflow.

Static analysis

No suspicious patterns detected.