guaikei·小红书数据

• Outcome: Review
• Latest audit: Aug 13, 2026 · 3:49 AM UTC
• Version: 1.0.0

Overview

This skill is a disclosed Xiaohongshu public-data collection tool that uses a third-party API and saves local result logs, with privacy considerations but no artifact-backed malicious behavior.

Install only if you are comfortable sending Xiaohongshu links, keywords, filter choices, and retrieved public-content data to guaikei.com using your GUAIKEI_API_TOKEN. Treat xsec_token-bearing URLs as potentially sensitive shared links, and periodically clean the local logs directory if the searches, comments, or competitor research should not remain on disk.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Description-Behavior Mismatch

Medium

• Confidence: 92% confidence
• Finding: The README asserts the tool only accesses public Xiaohongshu data, yet the examples require URLs containing xsec_token parameters. Even if those tokens are commonly present in shared links, they are still access-bearing parameters and their use conflicts with the privacy/scope claims, creating a risk of over-collection, mishandling of sensitive link tokens, or misleading users about what data is being processed.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill sends user-supplied Xiaohongshu links, search keywords, and fetched content through a third-party API (`guaikei.com`), but the main execution guidance does not require an upfront warning or user consent at the point of use. This creates a data-transmission/privacy risk because users may provide sensitive research targets, tracking subjects, or business-intelligence queries without realizing those inputs and resulting data are being relayed externally.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The README states that all task results are automatically saved to a local logs directory but does not clearly warn users about retention, sensitivity, or cleanup. Because the tool handles scraped content, comments, and token-bearing URLs, silent persistence can expose collected datasets or access-bearing parameters to other local users, backups, or later compromise.

Missing User Warnings

Low

• Confidence: 93% confidence
• Finding: The CLI writes full search results to a local JSON file containing the queried keyword and returned content metadata, but this file behavior is not clearly disclosed in the command's user-facing flow. In a data-collection skill, silent persistence can expose sensitive research terms or scraped content to other local users, backups, or downstream tooling that reads the workspace.

Missing User Warnings

Low

• Confidence: 88% confidence
• Finding: The script sends the user-provided keyword and filter parameters to remote search APIs, which is expected for this feature but is not clearly surfaced as a privacy-relevant action at runtime. Users may submit proprietary campaign terms, competitor names, or other sensitive research inputs without realizing they are being transmitted to an external service.

VirusTotal

64/64 vendors flagged this skill as clean.

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical

• Code: suspicious.exposed_secret_literal
• Location: SKILL.md:16