ClawHub
Back to skill

Security audit

Tavily Best Practices

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only Tavily integration guide; its API key, package install, search, extraction, and crawling examples match its stated purpose.

Reasonable to install as a reference guide. Before running examples, verify the Tavily packages, store API keys in environment variables or a secret manager, and set explicit limits and domain/path filters before crawling or saving page content.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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 (1)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill explicitly recommends omitting `chunks_per_source` to obtain full page content for data collection, but it does not warn that this may store large amounts of third-party content locally, including sensitive, personal, confidential, or copyrighted material. In a production best-practices skill for agentic workflows, that omission can lead users to implement bulk collection patterns without appropriate consent, filtering, retention, or compliance controls.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal