guaikei-rednote-scout

• Outcome: Review
• Latest audit: Aug 11, 2026 · 1:08 AM UTC
• Version: 1.0.0

Overview

This skill is a coherent Xiaohongshu public-data research tool, but users should understand that queries, URLs, API tokens, and fetched results are sent to guaikei.com and saved locally.

Install only if you are comfortable using a third-party guaikei.com API for Xiaohongshu research. Treat saved logs as potentially sensitive business or social-media research data, delete them when no longer needed, and avoid submitting private, restricted, or unauthorized content.

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
• 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

Findings (5)

Vague Triggers

Medium

• Confidence: 86% confidence
• Finding: The description says the skill applies even when the user does not mention operations, as long as the goal is to make decisions with Xiaohongshu data. That broad routing language can cause over-activation on generic planning or growth requests, leading the agent to invoke a third-party data integration unexpectedly and potentially send user-provided keywords or links off-platform without sufficiently specific user intent. In this context, the danger is moderate rather than severe because the skill is limited to public Xiaohongshu data and does not claim privileged account access, but unnecessary external calls and scope creep remain real risks.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The README states that all task results are automatically saved to a local `logs/` directory, but it does not clearly warn users that collected note, profile, and comment data may be persisted on disk. Because this skill is explicitly used for large-scale social-media data collection and comment analysis, silent local retention increases the risk of unintended exposure, over-retention, and mishandling of scraped personal or sensitive content.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The CLI persists fetched note details and comments to a local JSON file automatically, which can store potentially sensitive or regulated content on disk without clear opt-in or retention controls. In a data-collection skill focused on social-media intelligence, this increases the risk of unintended local exposure through shared machines, backups, CI logs, or later reuse of scraped data beyond the user's expectations.

Missing User Warnings

Medium

• Confidence: 80% confidence
• Finding: The tool sends a user-supplied Xiaohongshu URL together with an API token to backend task-creation and retrieval functions, but this file provides no explicit notice that note metadata/comments will be transmitted to an external service. While this is expected behavior for a network-backed data skill, the absence of transparent disclosure and consent increases privacy and compliance risk, especially when users may process competitor, KOL, or comment data at scale.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The CLI writes the full structured output to a local file, including the user-supplied profile URL and result data. Because Xiaohongshu profile links can contain query parameters such as xsec_token, this can persist sensitive request data to disk without an explicit warning, increasing the risk of local disclosure through shared machines, backups, or accidental log collection.

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