guaikei-douyin-data-getter

• Outcome: Review
• Latest audit: Aug 25, 2026 · 4:56 AM UTC
• Version: 1.0.0

Overview

The skill mostly matches its Douyin data-collection purpose, but it needs review because it can trigger on broad non-Douyin requests, sends user queries and URLs to a third-party API, stores fetched results locally, and exposes download-capable media fields despite saying downloads are out of scope.

Review before installing. Use this only when you intentionally want Douyin public-data collection through guaikei's API, and avoid sending sensitive market research terms or private URLs unless that is acceptable. Expect result JSON files to remain in the skill's logs directory, and treat bulk comment/account exports as data that may require privacy and platform-policy review.

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 (11)

Description-Behavior Mismatch

Medium

• Confidence: 83% confidence
• Finding: The schema explicitly documents `video_url` as including both playback and download addresses, which conflicts with the skill metadata stating it does not handle video downloading. That mismatch can enable downstream agents or integrators to use the field for downloading anyway, weakening policy boundaries and increasing the chance of unauthorized content acquisition or misuse.

Description-Behavior Mismatch

Medium

• Confidence: 93% confidence
• Finding: The schema explicitly documents `play_addr` as both a playback and video download address, which conflicts with the skill metadata stating it does not handle video downloading. This mismatch can enable downstream agents or users to discover and use a download-capable field anyway, weakening policy controls and potentially facilitating unauthorized content acquisition.

Description-Behavior Mismatch

Medium

• Confidence: 92% confidence
• Finding: The CLI writes full search output to a local file even though the skill is described as a retrieval/analysis tool and the user is not clearly informed that results will be persisted. Search results, keywords, and metadata may contain sensitive business research or user-derived content, creating unnecessary data retention and privacy risk if local files are later accessed by other users or processes.

Vague Triggers

High

• Confidence: 92% confidence
• Finding: The description says the skill should be used even when the user did not mention Douyin, for broad requests like competitor analysis or 'what is hot in the industry.' That can cause over-triggering in unrelated workflows, leading the agent to access external data sources or consume a sensitive API token when the user asked for general research, not Douyin-specific collection.

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The guidance to trigger during buried multi-step workflows is underspecified and encourages implicit activation without a clear platform boundary. In agent orchestration, this increases the chance of unintended tool use, unnecessary external API calls, and disclosure of user intent or query terms to a third-party service without sufficiently explicit user direction.

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The trigger phrases for activating the skill are broad enough to match ordinary user requests such as 'today what is hot' or generic 'search/find' language, which can cause the agent to invoke this external-data skill when the user did not explicitly intend Douyin data collection. That creates a prompt/skill-routing risk because user queries, search terms, or URLs may be sent to a third-party service without sufficiently clear consent or platform specificity.

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The post/account retrieval activation rules treat broad terms like '作品', '主页', '账号', or '博主' as sufficient to invoke account scraping behavior, while also warning that '视频' alone is ambiguous. In practice this can still overmatch common conversation about profiles or creators and trigger collection of account-level data from an external service without clear user intent, increasing privacy and unintended-action risk.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The README explains how to use the skill and obtain a token, but it does not clearly warn users that their search keywords, profile URLs, video URLs, and related query parameters will be transmitted to an external service operator. Because the skill is specifically designed for gathering third-party platform data, omission of this disclosure can mislead users about where their inputs go and create privacy, compliance, and trust issues.

Missing User Warnings

Medium

• Confidence: 87% confidence
• Finding: The document explicitly enables bulk collection of Douyin comment data (`--limit` up to 10000) but provides no warning about personal data, platform terms, consent, retention, or lawful use. In a social-media analytics skill focused on competitor analysis and comment mining, this omission increases the risk of privacy-invasive scraping, profiling, or downstream misuse of user-generated content.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The CLI persists fetched comment data to a local JSON file automatically after successful execution, without requiring explicit user consent or providing a clear privacy warning. Because comment data can contain personal information, account identifiers, or sensitive user-generated content, silent local retention increases the risk of unintended disclosure through shared machines, backups, log collection, or later exfiltration.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: Search results are silently written to disk after execution completes, with no clear warning or consent flow to the user. In this skill context, users may reasonably expect data retrieval only; silent persistence can expose competitive research, keywords, and collected content through leftover local artifacts.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.