ClawHub

Functions

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward Browserbase Functions deployment guide, but it requires deliberate handling of API keys, package installs, and cloud automation.

Install this only if you intend to let the agent help create and deploy Browserbase Functions. Review generated automation before publishing, run package commands in the intended project directory, keep .env files out of version control, avoid printing secrets in logs, and prefer scoped API keys, test accounts, or short-lived credentials for authenticated automation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly instructs users to copy Browserbase API credentials into a local .env file without any guidance on secure secret handling, exclusion from version control, or least-privilege usage. This creates a realistic risk of accidental credential disclosure through commits, logs, backups, or shared workspaces, especially in a deployment-oriented skill where users are likely to follow commands verbatim.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The authentication example demonstrates collecting and using email/password values from params but provides no warning about treating them as sensitive secrets, avoiding logs, or preferring token-based auth. In an automation/deployment skill, this normalizes passing raw credentials through invocation parameters, which may be retained in request history, logs, or debugging output.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal