guaikei-douyin-keyword-index-ranker

• Outcome: Review
• Latest audit: Aug 24, 2026 · 3:24 AM UTC
• Version: 1.0.0

Overview

The skill mostly performs Douyin public-data collection as advertised, but it needs review because it can run on generic short-video research requests and saves collected results locally by default.

Review before installing. Use it only when you intend Douyin data collection, keep the GUAIKEI_API_TOKEN private, and remember that collected search/post/comment results may be saved under the skill's logs directory by default. Delete or protect those logs if they contain user-generated comments, account metadata, or business research.

SkillSpector

By NVIDIA

Vulnerability Patterns

• 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
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Description-Behavior Mismatch

Medium

• Confidence: 93% confidence
• Finding: The CLI writes fetched Douyin comment data to a local JSON file after printing results, but this persistence is not surfaced as an explicit user-consented behavior in this file. Comments may contain personal data or sensitive business research artifacts, so silently storing them increases exposure through leftover files, accidental sharing, or collection by other local users/processes.

Vague Triggers

High

• Confidence: 95% confidence
• Finding: The description says the skill should activate for broad short-video operations and multi-step workflows even when the user did not mention Douyin. That can cause overbroad invocation, collecting external platform data without clear user intent and unexpectedly expanding task scope, which is a prompt-routing and consent problem.

Vague Triggers

High

• Confidence: 96% confidence
• Finding: The generalized trigger section explicitly instructs use for broad research, competitor analysis, and topic selection tasks even without clear platform scope. In context, this increases the chance of unintended data collection and tool execution, especially in workflows where the user asked for generic research rather than Douyin-specific scraping.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The README states that structured JSON logs are automatically written locally, but the user warning is weak and fragmented rather than clearly foregrounding that scraped public data, including comments and account/video metadata, will be exported to disk by default. In a data-collection skill, silent or poorly disclosed local retention increases privacy, compliance, and accidental exposure risk because operators may handle regulated or sensitive content without realizing it is being stored persistently.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The changelog explicitly advertises bulk collection of competitor account content and retrieval of video comments, but provides no warning about privacy, platform terms, retention, or downstream handling of personal data. In this skill context, the feature is designed for large-scale analytics workflows, which increases the chance of misuse, over-collection, or non-compliant processing of user-generated content.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: This code saves the full comment retrieval output to a local file without any confirmation or visible notice beyond the source code itself. In the context of a research/analysis skill that processes public comments at scale, silent retention can create privacy, compliance, and operational security issues because users may assume the tool only fetches/analyzes data transiently.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.