Security audit

guaikei-douyin-data-intake-to-json

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent Douyin public-data collection tool, but it needs Review because it can send broad user queries and URLs to a third-party API and automatically store large comment/account datasets locally.

Install only if you are comfortable sending Douyin search terms, account/video URLs, request limits, and your GUAIKEI_API_TOKEN to GuaiKei, and with results being saved automatically in the skill's logs directory. Avoid sensitive internal research terms, review/delete stored logs, and ensure collection of public comments and account data complies with platform terms and applicable privacy rules.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (7)

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The generalized trigger says the skill should activate even when the user does not explicitly mention Douyin, including in broad research workflows. That can cause unintended invocation and collection of third-party platform data, which is risky because this skill also performs network access and automatically logs outputs to disk.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill states that output files are automatically saved to `logs/`, but the description and invocation guidance do not prominently warn users before collection begins. Silent persistence of collected comments, account data, and search results can expose sensitive business research or personal data to other local users, backup systems, or downstream tooling.

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The natural-language trigger table maps very broad phrases such as '搜索 / 搜一下 / 找 AI 相关视频' into executable commands, which can cause unintended invocation when a user speaks casually or ambiguously. In an agent setting, overbroad triggers increase the chance of the skill being selected and sending user-supplied queries or URLs to the external GuaiKei service without sufficiently explicit user intent.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The intent recognition priority uses generic terms like '热搜', '热点', '榜单', and '今天什么火' as direct routing signals to the hot-list command, without enough context constraints. This makes accidental activation more likely in unrelated conversations about trends, causing external data requests and potentially confusing or privacy-impacting behavior in an autonomous agent workflow.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The README tells users to obtain and configure an API token but does not clearly warn that search keywords, URLs, and related request data are transmitted to an external third-party service. This lack of transparency can lead users or downstream agents to expose sensitive queries, monitored account URLs, or workflow context to the service without informed consent.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The documentation explicitly guides users to collect Douyin creator posts and video comments at scale, but provides no notice about privacy, consent, retention, or legal/compliance constraints. In this skill’s context, that omission increases the likelihood of indiscriminate collection of personal data and user-generated content, which can lead to privacy misuse or regulatory exposure even if the data is publicly accessible.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The CLI persistently writes fetched comment data to a local JSON file automatically, but the code provides no explicit user-facing consent or warning that potentially sensitive scraped content will be stored on disk. In shared environments, CI runners, or multi-user systems, these files may be retained, indexed, backed up, or accessed by others, increasing privacy and data-handling risk.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.