guaikei-douyin-trending-for-hotspot-tracking

• Outcome: Review
• Latest audit: Aug 22, 2026 · 3:38 AM UTC
• Version: 1.0.0

Overview

The skill mostly matches its Douyin research purpose, but it can run from broad prompts, send targets to a third-party API, and save full results locally without per-run consent.

Install only if you are comfortable with Douyin research prompts being sent to guaikei.com with your API token and with results being retained in local log files. For ambiguous content-planning requests, confirm the user really wants Douyin data before running the skill, and delete logs when they contain sensitive research.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (8)

Vague Triggers

High

• Confidence: 93% confidence
• Finding: The skill instructs activation even when the user does not mention Douyin explicitly, based on broad intent like content research or topic planning. This can cause over-triggering in unrelated workflows, leading the agent to invoke third-party data collection unexpectedly and potentially send user queries or derived business context to an external service without clear user intent.

Vague Triggers

Medium

• Confidence: 89% confidence
• Finding: Example triggers such as generic search-style requests are not sufficiently constrained to the Douyin platform, which increases the chance the agent routes ordinary search tasks into this skill. In context, this matters because the skill performs external data retrieval and token-authenticated API usage, so mistaken invocation can leak query intent and produce irrelevant or privacy-impacting collection.

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The natural-language trigger examples are broad enough to activate on generic user phrasing such as '找' or '视频', increasing the chance that the skill runs in contexts where the user did not explicitly intend Douyin scraping. In a multi-skill agent, this can cause overcollection of public platform data, unintended external API calls, and disk writes to logs without sufficiently clear user consent.

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The intent recognition priority rules define activation based on very common terms like '热点', '搜索', and '找 + 关键词' without enough scope constraints, which can cause the skill to hijack unrelated requests embedded in broader workflows. Because this skill performs scraping-oriented retrieval and automatic logging, erroneous activation materially increases privacy, compliance, and least-astonishment risks.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The README emphasizes data export and logging but does not prominently warn that scraped public data and user-supplied targets will be automatically written to local log files. This can lead users or downstream agents to collect and persist potentially sensitive comment data or competitive intelligence on disk without informed awareness, increasing exposure through retention and accidental sharing.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The CLI persists the full retrieved results to a local JSON file automatically, without any opt-in, warning, or indication of where the data is stored. Because this skill handles scraped Douyin account and content data, silent local retention can expose potentially sensitive research data, comments, or account intelligence to other local users, backups, or downstream tooling that reads the filesystem.

Missing User Warnings

Medium

• Confidence: 87% confidence
• Finding: The code sends user-supplied URL and limit values to remote API functions without clearly disclosing in this file that the inputs will be transmitted off-host. In a data-collection skill that may be invoked implicitly inside broader workflows, this creates a meaningful privacy and consent risk because users may not realize account identifiers or research targets are being shared with an external service.

Missing User Warnings

Low

• Confidence: 93% confidence
• Finding: The CLI persists full search results to a local JSON file without clearly informing the user that their query and returned data will be stored on disk. In this skill's context, the results may include operational research data, account analysis, or comment data that users may reasonably expect to be ephemeral, creating a privacy and data-retention risk on shared machines or managed environments.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.