Auto Search using Google Baidu

• Outcome: Pass
• Latest audit: May 29, 2026 · 8:24 PM UTC
• Version: 1.5.0

Overview

This skill is a disclosed Google/Baidu web-search helper; its external queries and API-key use fit its purpose, with some privacy and dependency hygiene caveats.

Install only if you are comfortable sending search terms to Google or Baidu under the API keys you configure. Avoid searching for secrets or confidential data, use limited-scope or quota-limited provider keys, and consider pinning dependencies or installing in an isolated environment for sensitive use.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (11)

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The README promotes automatic querying of Google and Baidu but does not disclose that user-entered search terms will be sent to external third-party providers. This creates a real privacy and consent issue, especially because search terms may contain sensitive personal, corporate, or regulated data and are routed to different providers depending on content.

Natural-Language Policy Violations

Medium

• Confidence: 82% confidence
• Finding: The skill description states that engine selection is performed automatically based on query language, with Chinese queries routed to Baidu and others to Google, but no user opt-in or preference control is described. This can misroute sensitive queries across jurisdictions or to providers a user would not choose, creating a privacy and policy risk tied to language-based routing.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill documentation describes sending user queries to Google and/or Baidu but does not clearly warn that search terms are transmitted to third-party providers and may be logged, profiled, or processed under those providers' policies. Because search queries can contain sensitive personal, corporate, or regulated information, the lack of an explicit privacy warning can cause unintended data disclosure by users or downstream agents.

Natural-Language Policy Violations

Medium

• Confidence: 88% confidence
• Finding: The documented behavior automatically routes queries to Google or Baidu based on language/content without requiring user opt-in, which can silently disclose user input to a provider they did not choose. This is especially sensitive because language-based routing may reveal nationality, language preference, or topic context and may direct data into different legal jurisdictions and surveillance regimes.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The code sends user search queries directly to Google over the network, which discloses potentially sensitive user input to a third-party provider without any explicit consent flow or warning in this skill. In an agent setting, users may reasonably assume local processing, so silent transmission of prompts, names, identifiers, or proprietary terms creates a real privacy and compliance risk.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The Baidu search path transmits user queries to an external third party without an explicit warning or opt-in, creating a privacy disclosure issue similar to the Google path. This is somewhat more sensitive in context because the code automatically routes Chinese- or China-related queries toward Baidu, potentially sending user input to a provider in a different jurisdiction without the user realizing it.

Natural-Language Policy Violations

Medium

• Confidence: 95% confidence
• Finding: The skill automatically classifies queries by language/China-related keywords and routes them to different providers without user opt-in, removing meaningful user control over where their data is sent. This increases privacy risk because routing decisions are based on query content itself, and users may not expect politically, geographically, or linguistically inferred data to determine third-party disclosure.

Unpinned Dependencies

Low

• Category: Supply Chain

Content

requests>=2.28.0
python-dotenv>=1.0.0

• Confidence: 95% confidence
• Finding: requests>=2.28.0

Unpinned Dependencies

Low

• Category: Supply Chain

Content

requests>=2.28.0
python-dotenv>=1.0.0

• Confidence: 93% confidence
• Finding: python-dotenv>=1.0.0

Known Vulnerable Dependency: requests — 10 advisory(ies): CVE-2014-1830 (Exposure of Sensitive Information to an Unauthorized Actor in Requests); CVE-2024-47081 (Requests vulnerable to .netrc credentials leak via malicious URLs); CVE-2024-35195 (Requests `Session` object does not verify requests after making first request wi) +7 more

High

• Category: Supply Chain
• Confidence: 98% confidence
• Finding: requests

Known Vulnerable Dependency: python-dotenv — 1 advisory(ies): CVE-2026-28684 (python-dotenv: Symlink following in set_key allows arbitrary file overwrite via )

Low

• Category: Supply Chain
• Confidence: 72% confidence
• Finding: python-dotenv

VirusTotal

64/64 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.