ClawHub
Back to skill

Security audit

PrestaShop Bridge V1

Security checks across malware telemetry and agentic risk

Overview

This looks like a coherent PrestaShop bridge skill, but it needs review because it can change live store data and ships an unlabeled signing secret in its validator.

Install only if you control the PrestaShop bridge and can enforce least-privilege credentials. Do not reuse the embedded HMAC value; rotate or reject it if it was ever trusted. Require explicit confirmation for product imports, price or stock changes, SEO edits, and order-status updates, and verify the publisher/provenance before using this in production.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Intent-Code Divergence

High
Confidence
99% confidence
Finding
The file hardcodes a long-lived HMAC secret directly in source and uses it to generate request signatures. Even though this is presented as a validator for example payloads, embedding a real signing key in a distributed skill undermines the documented security model of environment-managed secrets and allows anyone with the skill contents to mint valid signatures for the covered API contract if the same secret is accepted by the bridge.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The code silently uses an embedded secret to compute exact HMACs for sample requests, with no indication that the secret is test-only or segregated from production. In a skill explicitly marketed as secure and signed, this context makes the practice more dangerous because users may assume the examples reflect safe operational patterns while the source actually discloses signing material.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.