ClawHub
Back to skill

Security audit

tanstack

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only TanStack skill, but some examples under-explain privileged CDN operations, client-exposed environment values, and public URL data, so it belongs in Review rather than automatic install.

Review generated code before applying this skill’s examples. Do not copy the CDN purge endpoint without stronger authentication, path allowlisting, rate limiting, and audit logging. Treat all URL search params, including Base64-encoded ones, as public. Only return allowlisted non-sensitive runtime config to clients, and verify TanStack Query v5 API advice before using infinite-query refetch examples.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Intent-Code Divergence

Medium
Confidence
98% confidence
Finding
The document explicitly states earlier that TanStack Query v5 removed `refetchPage`, but the Best Practices section later recommends using it. This contradiction can mislead users into relying on removed or unsupported API behavior, causing broken implementations or unsafe assumptions about how selective refetching works.

Context-Inappropriate Capability

Medium
Confidence
82% confidence
Finding
The documentation includes a credentialed outbound request to a third-party CDN purge API using powerful environment-backed tokens. In a skill intended for app-building guidance, this broadens capability into infrastructure mutation and could be copied into projects without sufficient guardrails, enabling cache purge abuse, unauthorized operational changes, or accidental exposure of sensitive operational workflows.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The documentation recommends Base64 encoding for URL search params as a compatibility technique, but it does not warn that Base64 is only encoding, not protection. Developers may incorrectly place sensitive values into query strings, which remain exposed in browser history, server logs, analytics, referrer headers, bookmarks, and shared links, and are trivially decoded by anyone who sees the URL.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The example returns a server environment variable directly to the client, while describing it generically as a runtime variable. Without an explicit warning, users may copy the pattern for secrets, API keys, or internal configuration, causing server-only values to be serialized into loader data and exposed to browsers.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.