Security audit

guaikei-xhs-search-and-track

Security checks across malware telemetry and agentic risk

Overview

This skill coherently retrieves public Xiaohongshu data through a disclosed third-party API, but users should understand that results are saved locally by default.

Install only if you are comfortable sending Xiaohongshu keywords or URLs to Guaikei's API using your GUAIKEI_API_TOKEN. Treat the generated logs as collected social-media datasets: keep them out of shared or synced directories if needed, and delete them when no longer required.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (5)

Context-Inappropriate Capability

Medium

Confidence: 90% confidence
Finding: The CLI persists the full fetched result set, including note and comment data, to a local file after execution. For a skill described as read-only retrieval of public Xiaohongshu data, silently creating local data archives expands data retention and exposure beyond the immediate request, increasing the risk of unintended disclosure through shared hosts, backups, or later access by other users/processes.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The README states that all task results are automatically saved to the `logs/` directory, but it does not clearly warn users that fetched note details, comments, and other scraped public-content datasets may be written to local files. In a data-collection skill, silent persistence increases the risk of unintended retention, redistribution, or exposure of potentially sensitive or regulated content, especially on shared machines or in synced workspaces.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The CLI writes the full task output to a local JSON file named from the target URL without any explicit user warning or opt-in. Even though the data is sourced from public Xiaohongshu content, the saved file may contain scraped profile, post, or comment data that persists on disk and can be exposed to other local users, backups, or later unintended reuse.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The supplied Xiaohongshu URL and requested limit are sent to an external API through createPostTask/getPostTask, but this file does not clearly disclose that user-provided data is transmitted off-host. Users may assume the skill operates locally, leading to unintended sharing of investigation targets, tokens embedded in URLs, or other sensitive query context with a third-party service.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The CLI writes a log file whose filename includes the user-provided keyword and whose contents include the full structured search results. This can expose potentially sensitive research terms, monitoring targets, or collected third-party data to other local users, backups, or log collectors without any explicit warning or consent in this file.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.