guaikei-douyin-quick-fetch-public

• Outcome: Review
• Latest audit: Aug 24, 2026 · 4:47 AM UTC
• Version: 1.0.0

Overview

The skill largely matches its stated Douyin public-data purpose, but its trigger scope is too broad and it automatically stores scraped search, profile, and comment data locally.

Review before installing. Use it only when you intentionally want Douyin public-data collection through guaikei.com, and be aware that successful search, creator-post, and comment fetches are saved locally under logs. Avoid using it for generic research prompts unless Douyin is explicitly the intended platform, and treat saved comment/profile data as potentially personal data.

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 (9)

Description-Behavior Mismatch

Medium

• Confidence: 93% confidence
• Finding: The CLI saves full search output to a local JSON log file after completing what is presented as a fetch/read-only operation. This creates unexpected persistent storage of potentially sensitive query terms and returned data, which can expose user activity or collected platform data to other local users, backups, or later reuse without the user's awareness.

Vague Triggers

High

• Confidence: 94% confidence
• Finding: The skill is configured to activate even when users do not mention Douyin, based on broad research-oriented language. This can cause overbroad invocation, unexpected external data access, and accidental transmission of user queries to a third-party API in contexts where the user did not clearly consent to using this specific platform integration.

Vague Triggers

High

• Confidence: 95% confidence
• Finding: The generalized activation guidance explicitly tells the agent to trigger on vague multi-step requests like generic sector research. In an agentic environment, this increases the risk of tool mis-selection and unnecessary disclosure of user prompts, URLs, or research targets to an external service without sufficiently specific user authorization.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The natural-language trigger phrases are broad enough to match common user requests outside a clearly scoped Douyin-data context, which can cause the agent to invoke this skill unexpectedly. Because the skill performs external data collection and writes logs locally, mistaken invocation can lead to unintended scraping actions, unnecessary token use, and collection/storage of data the user did not specifically request.

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The intent-recognition priority rules use ambiguous terms such as '热搜', '搜索', '找', '作品', and '视频' with insufficient scope constraints, increasing the chance that unrelated user requests will be routed into this scraping skill. In an agent setting, this can produce unauthorized or surprising external actions, especially since the README states the skill may be used even when the word '抖音' is not mentioned.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The README promotes exporting scraped video, author, comment, and ranking data to local JSON logs but does not give meaningful privacy, retention, access-control, or lawful-use guidance. Even if the data is described as public, comments and account-related data can still contain personal information, and silent local storage increases the risk of overcollection, improper sharing, or noncompliant downstream use.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The CLI persists the full comment fetch output to a local JSON file via log.taskWrite, and that output likely contains user-generated comments and associated metadata. Storing this data silently increases exposure through local filesystem access, backups, shared workstations, CI artifacts, or accidental commit/upload, especially because users are not clearly warned that data will be retained.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The CLI writes the full fetched results object to a local JSON file automatically, with no opt-in, warning, or redaction step in this file. Because the skill is designed to collect Douyin profile/content data for operations, competitive analysis, and public-opinion workflows, the saved file may contain scraped identifiers, profile metadata, and content that can persist on disk longer than intended and be exposed to other local users, backups, or later processes.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The program writes search results to a local log file without explicit user warning or consent, despite behaving like a data retrieval tool. Silent persistence increases privacy and data-governance risk because search queries and results may remain on disk indefinitely and be accessible outside the immediate session.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.