travily_search_pro

• Outcome: Review
• Latest audit: May 29, 2026 at 9:25 PM UTC
• Version:
  1.0.1

Overview

The skill mostly does what it claims, but it bundles and automatically loads a Tavily API-key-like value while sending user queries and URLs to an external service.

Review before installing. Replace or remove the bundled .env value and use your own Tavily API key, avoid submitting secrets, private/internal URLs, customer data, or regulated content to Tavily, and choose --output paths carefully because existing files may be overwritten.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 93% confidence
• Finding: The skill documentation describes capabilities that use environment variables, write output files, and make outbound network requests, but it does not declare corresponding permissions. This can mislead users and security controls about what the skill is able to access, reducing transparency and making unintended data exposure or file modification more likely.

Context-Inappropriate Capability

Medium

• Confidence: 87% confidence
• Finding: The script accepts an arbitrary --output path and writes fetched content directly to that location without validation or confirmation. In agent or automated contexts, this can overwrite local files unexpectedly, which expands the skill from network extraction into local file modification.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill advertises search, research, and URL extraction against Tavily but does not clearly warn that user queries, research topics, and supplied URLs will be transmitted to an external third-party service. Users may unknowingly send sensitive prompts, internal URLs, or proprietary research topics off-platform, creating confidentiality and compliance risks.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: A user-supplied URL is transmitted to Tavily's external API, which may disclose private or internal URLs, tokens embedded in query strings, or other sensitive references to a third party. This is especially relevant in agent settings where users may assume local-only processing.

Missing User Warnings

Low

• Confidence: 84% confidence
• Finding: Writing extracted content to a file without warning can lead to accidental overwrite or unintended local state changes. While not inherently malicious, it is a meaningful safety issue for a tool whose primary purpose is content extraction rather than file management.

Missing User Warnings

Medium

• Confidence: 84% confidence
• Finding: The script sends the user-supplied research topic directly to an external Tavily service without any explicit disclosure prompt or warning at execution time. If users enter sensitive internal data, credentials, customer information, or confidential prompts, that data will be transmitted off-host to a third party. In an agent-skill context, this is more dangerous because the skill may be invoked as part of automated workflows where users may not realize external sharing occurs.

VirusTotal

64/64 vendors flagged this skill as clean.