ClawHub
Back to skill

Security audit

Price Import

Security checks for vulnerabilities and agentic risk

Overview

This skill mostly imports prices, but it also includes broader product-changing and requirements-management workflows that need review before installation.

Install only if you are comfortable giving the agent controlled API authority to import prices, upload source files, and potentially affect product/catalog workflows. Before use, clarify whether product creation is allowed, restrict or remove the /fix workflow if you only need price import, and review every import or generated requirement document before confirming.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (3)

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The skill is declared as a price-import tool, but it also exposes a separate /fix workflow for requirements intake and task orchestration. This scope expansion increases the attack surface and can let users trigger unrelated operational actions under the guise of a simple import command set, violating least privilege and creating confusion about what the skill is allowed to do.

Context-Inappropriate Capability

High
Confidence
97% confidence
Finding
The embedded software change-request and developer task-management capability is unrelated to price importing and includes generation of requirement documents, registry updates, and coordination with another agent. In an agentic environment, this creates a privilege and scope escalation path from data import into persistent workflow automation and local file manipulation, which can be abused through prompt-driven requests.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill describes writing and updating local files such as matching-rules.json and later requirement documents without consistently warning the user that persistent workspace state will be created or modified. Silent persistence is dangerous because attacker-controlled content can be stored and later influence future runs, leading to configuration poisoning, unexpected behavior, or data leakage across sessions.

Static analysis

No suspicious patterns detected.