Back to skill

Security audit

Marktplaats

Security checks across malware telemetry and agentic risk

Overview

The skill mostly matches a Marktplaats selling assistant, but its Safari probing can inspect logged-in browser pages and user-supplied URLs without an enforced Marktplaats-only scope.

Install only if you are comfortable with a local assistant using your logged-in Safari session for Marktplaats. Before running probes, keep Safari on the intended Marktplaats page, avoid arbitrary --url or --open-background targets, and treat saved ad records/snapshots as private because they may contain listing details and form-derived data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding
The skill advertises capabilities implying shell and network access but does not declare permissions, which weakens transparency and security review. Undeclared execution and outbound connectivity can enable unexpected command execution or data transfer paths that users and policy enforcement may not anticipate.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding
The declared description presents a narrow publishing update, but the actual behavior includes broad live-site interaction, browser automation, HTTP probing, local file processing, and register updates. This mismatch is dangerous because it can conceal high-risk capabilities from users and reviewers, increasing the chance of unintended data access, unsafe automation, or misuse of authenticated browser context.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The `--browser-fetch` path performs an in-browser same-session request with `withCredentials = true`, causing Safari's authenticated cookies to be sent to the target URL automatically. Because the script also supports a user-provided `--url` and can open pages in the background, it risks silently querying authenticated endpoints and exposing account-specific form data or other sensitive page contents without an explicit confirmation step.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.