Hybrid Deep Search

• Outcome: Pass
• Latest audit: May 28, 2026 at 6:16 AM UTC
• Version:
  1.0.1

Overview

This appears to be a user-directed search helper with some privacy and reliability caveats, but no evidence of hidden exfiltration, persistence, or destructive behavior.

Install only if you are comfortable with an experimental search helper. Avoid putting secrets, personal data, internal URLs, or proprietary text in queries; prefer environment variables for API keys; monitor OpenAI costs; and verify real provider integration before relying on returned search results.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (4)

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill auto-routes user queries to external providers, including OpenAI and Brave, but does not prominently warn users that their prompts may be transmitted off-platform. This is dangerous because users may submit sensitive data under the assumption the query stays local, leading to unintended disclosure to third parties.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The documentation shows storing an API key directly in config.json without warning about secret handling or accidental commit risk. This encourages insecure credential storage practices that can easily lead to key leakage through source control, backups, logs, or file sharing.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill routes user queries to external services including Brave and OpenAI, but the usage section does not clearly warn that user-supplied content may leave the local environment and may incur charges. This creates a real privacy and cost-risk because users may submit sensitive prompts assuming the skill is local-only or free-by-default.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The report documents searches being sent to Brave API and OpenAI/Codex but does not warn users that their queries may be transmitted to third-party services, potentially exposing sensitive prompts, internal data, or personal information. In a search skill that automatically routes queries externally, the absence of a privacy disclosure increases the chance that users unknowingly send sensitive content off-system.

VirusTotal

66/66 vendors flagged this skill as clean.