guaikei-douyin-full-scan-posts

• Outcome: Review
• Latest audit: Aug 20, 2026 · 2:49 AM UTC
• Version: 1.0.0

Overview

This is a real Douyin public-data collection skill, but it can over-trigger broad requests and saves high-volume scraped results locally by default.

Install only if you intend to send Douyin search terms, account/video URLs, and retrieval limits to Guaikei's API and to keep scraped results on disk. Use a dedicated token, avoid collecting more comments or account data than needed, delete logs when finished, and confirm ambiguous requests before letting the skill run.

SkillSpector

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 (10)

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The CLI writes the full comment retrieval output, including all returned comments and request metadata, to a local JSON file by default. This creates a data exposure risk because comment content may contain personal data or sensitive content, and storing it persistently without opt-in, redaction, retention controls, or access restrictions increases the chance of unintended disclosure on shared systems or through later compromise.

Description-Behavior Mismatch

Low

• Confidence: 90% confidence
• Finding: The CLI persists fetched Douyin post results to a local JSON file after completing the request, but this storage behavior is not apparent from the stated user-facing capability of simply retrieving a creator's recent posts. Silent local retention can expose scraped content, query targets, and metadata to other local users, backup systems, or later unintended processing, especially if the results include sensitive or regulated content.

Description-Behavior Mismatch

Low

• Confidence: 91% confidence
• Finding: The CLI persists full search output to a local JSON log file, including the user's keyword and returned content, without any visible consent, retention control, or minimization. Even though this is framed as a Douyin search utility, silent local storage expands the data exposure surface and can leak sensitive search interests or collected content to other local users, backup systems, or later compromise.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The implicit trigger examples are broad enough to match ordinary research or trend-analysis requests such as '最近什么热门' or '竞品调研', which can cause the skill to activate when the user did not specifically ask for Douyin data collection. Because this skill can perform large-scale scraping of public content and comments, over-triggering increases the risk of unintended data access, unnecessary token use, and execution of networked collection actions outside user intent.

Missing User Warnings

Medium

• Confidence: 82% confidence
• Finding: The README promotes exporting scraped comments, author, and hot-list data to JSON for analysis, but it does not give clear privacy, retention, minimization, or consent guidance. Even if the source is described as public Douyin data, comment content and account-associated metadata can still contain personal data, and storing it by default increases the risk of unnecessary collection, internal misuse, or secondary redistribution.

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The natural-language trigger mapping uses broad conversational phrases and direct command templates that can cause the skill to activate on ambiguous user requests. In a data-scraping skill, unintended invocation can trigger collection of third-party content or comments the user did not clearly request, creating privacy, compliance, and operational risk.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The intent recognition priority rules rely on generic keywords such as hot topics, search, comments, homepage, or blogger with limited contextual safeguards. Because this skill performs large-scale Douyin data retrieval, overbroad matching makes accidental execution more dangerous than in a low-risk utility, potentially causing unauthorized or unexpected scraping actions.

Missing User Warnings

Medium

• Confidence: 83% confidence
• Finding: The tool sends a user-supplied Douyin URL and retrieval parameters to an external API using a bearer token, but provides no explicit user-facing notice or consent mechanism before transmitting that data off-host. In the context of a comment-fetching skill, this matters because queried URLs and associated comments may reveal user interests, targets of investigation, or other sensitive operational context, and the skill additionally persists results locally, increasing privacy risk.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The code writes fetched results to a local JSON file without any user-facing warning or consent prompt in this file. This creates a quiet data-retention channel for scraped content and request metadata, increasing the chance of privacy leakage, unauthorized reuse, or forensic recovery from disk beyond the user's immediate expectation of a transient lookup.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The code writes complete search results to a local JSON file without warning or user confirmation, creating an undisclosed persistence channel. This is particularly risky because search queries and returned Douyin content may reveal user interests, investigations, or other sensitive context, and the skill description does not prepare the user for local archival of that data.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.