Security audit

Baidu Intelligent Search - 百度智能搜索生成

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward Baidu search-and-answer integration, but users should remember that their queries are sent to Baidu under their API key.

Install this only if you are comfortable sending search questions and optional advanced request fields to Baidu Qianfan using your BAIDU_API_KEY. Avoid including secrets, private documents, or sensitive personal data in queries unless you intend Baidu to process that content, and review any raw_body payload before use.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The trigger description uses broad phrases like '搜一下并给个结论' and similar search-plus-summary requests, which can overlap with ordinary user conversation and cause overbroad invocation. Mis-triggering a networked skill can unintentionally send user prompts to an external provider, creating privacy and policy risks even without malicious intent.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The documentation states that the skill performs online search and model summarization through Baidu’s API, but it does not explicitly warn that user queries and possibly related context will be transmitted to an external third-party service. This creates a real transparency and privacy issue because users may disclose sensitive information without informed consent.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.