Guaikei Douyin Quick Fetch Public

• Outcome: Review
• Latest audit: Aug 25, 2026 · 4:09 AM UTC
• Version: 1.0.0

Overview

This skill performs the advertised Douyin data collection, but it is broad-triggered and automatically stores large public-profile/comment datasets locally without strong user control or retention guidance.

Install only if you are comfortable sending Douyin keywords, target URLs, limits, and a guaikei.com API token to that provider, and with fetched public user/comment data being saved under the skill's logs directory. For ambiguous research tasks, confirm the user explicitly wants Douyin collection before running it, and periodically delete or protect the generated logs.

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 (6)

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 91% confidence
• Finding: The skill claims a narrower, cleaner behavior profile than what is described elsewhere in the file: it persists fetched results to `logs/`, advertises '用户搜索' and '评论分析' capabilities that are not clearly implemented, and mixes collection with local retention. This mismatch is dangerous because users and orchestrators may grant trust or route sensitive research tasks based on inaccurate expectations, leading to unintended data retention and overestimation of analysis capabilities.

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The activation guidance is broad enough to trigger on generic requests like '帮我调研一下这个赛道' or any vague short-video research task even when the user did not request Douyin-specific data collection. Over-broad triggering is risky because it can cause unintended third-party data access, unnecessary token use, and collection/storage of external content outside the user's explicit intent.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The README explicitly states that fetched Douyin data and comment content are automatically exported to local JSON log files, but it does not give a clear privacy/security warning about storing potentially sensitive public-profile and comment data on disk. In a skill focused on large-scale collection, competitor monitoring, and sentiment analysis, silent persistence increases the risk of unintended retention, local exposure, and secondary misuse of scraped data.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The CLI writes fetched comment data to a local JSON file automatically, without informing the user or requiring opt-in. Because comment data may contain personal or sensitive content and the filename is predictable, this creates an avoidable local data-retention/privacy risk, especially on shared machines or in automated environments where logs and artifacts persist.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The tool sends the user-supplied URL and requested comment limit to external APIs to create and retrieve comment tasks, but it does not clearly disclose this network transmission at the point of use. In a privacy-sensitive context, users may not realize that queried targets and analysis parameters are being shared with a third-party service, which can expose research interests, monitoring targets, or other sensitive operational metadata.

Missing User Warnings

Low

• Confidence: 89% confidence
• Finding: The CLI automatically persists fetched Douyin results to a local JSON file after successful execution, but this file does not clearly warn the user or require opt-in consent before writing data to disk. Even though the source is public-platform data, the results may still contain profile identifiers, content metadata, or comments that create privacy, compliance, or accidental data-retention risks on shared systems.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.