ClawHub

Interview Buddy

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed third-party interview-coaching API skill, with privacy and billing considerations but no evidence of hidden, destructive, or deceptive behavior.

Install this only if you intend to use ToolWeb's hosted interview-coaching service. Protect the API key, expect successful calls to count against quota or billing, and avoid sending confidential employer, customer, or highly personal details unless you are comfortable with ToolWeb processing them.

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)

Context-Inappropriate Capability

Medium
Confidence
93% confidence
Finding
The skill introduces OTP-based authentication and session handling for a simple interview-practice workflow, causing collection of user email and one-time codes by an external service. This expands the trust boundary and data exposure beyond what users would reasonably expect from a coaching skill, creating privacy and account-handling risk if the external service is misused, compromised, or poorly explained.

Intent-Code Divergence

Low
Confidence
75% confidence
Finding
The documentation claims multiple endpoint-specific operations, but every example posts to the same generic URL without showing the listed paths, making the actual backend behavior opaque. This ambiguity hinders meaningful review of what data is being sent where and can conceal overbroad server-side actions behind a single endpoint.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill does not prominently warn users that their interview prompts, answers, email address, and OTP-related data are transmitted to a third-party API. Interview practice often includes sensitive career history, employer names, project details, and personal contact information, so undisclosed transmission materially increases privacy and confidentiality risk.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The skill requires an environment-stored API key and instructs the agent to always use it, but does not clearly warn users that requests are authenticated to a third-party service using deployer-controlled credentials. This can obscure who is authorizing the transmission, who is billed, and whether user content is being sent under shared credentials.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal