guaikei-douyin-intelligence

• Outcome: Review
• Latest audit: Aug 19, 2026 · 6:02 AM UTC
• Version: 1.0.0

Overview

The skill appears purpose-built for Douyin public-data intelligence, but it needs review because it can send broad user research requests to a third-party service and automatically stores collected results locally.

Review before installing if you will research sensitive competitors, accounts, URLs, or topics. Expect requests and target identifiers to be sent to guaikei.com and successful search/post/comment results to be saved under the skill's logs directory by default; protect the API token and periodically delete logs you no longer need.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (11)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 87% confidence
• Finding: The skill states it requires an environment token (`GUAIKEI_API_TOKEN`) and operationally depends on env access, but no explicit permission declaration is shown in the skill manifest/frontmatter. That creates a transparency and least-privilege problem: agents or reviewers may not understand that sensitive environment data is needed, increasing the risk of unsafe deployment or accidental token exposure through surrounding tooling.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The description says the skill should trigger even when the user does not mention Douyin, based only on broad competitor-analysis or industry-keyword requests. This can cause the agent to invoke external data collection unexpectedly, sending user queries to a third-party service without clear user intent or platform-specific consent.

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The implicit trigger examples are generic phrases like competitive research, high-like content, and hot topics, which overlap with common requests that may have nothing to do with Douyin. This broad matching raises the chance of unintended tool use, unnecessary data transmission, and surprising behavior for users who did not request this platform or external scraping workflow.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill advertises collection of public Douyin data, comments, and structured output, and later mentions automatic archival to `logs/`, but it does not prominently warn users that their requested targets, query terms, and retrieved content will be sent to an external service and stored locally. This is a disclosure and consent gap that can expose sensitive research interests or collected personal data from comments.

Missing User Warnings

Medium

• Confidence: 87% confidence
• Finding: The README promotes fetching public Douyin videos, authors, comments, and hot-list data and states that JSON logs are automatically saved locally, but it does not clearly warn users that comments, profile identifiers, and other retrieved fields may constitute personal data. This can lead users to collect and retain personal data without understanding privacy, compliance, and local storage risks, especially in competitor monitoring and sentiment analysis workflows.

Missing User Warnings

Low

• Confidence: 81% confidence
• Finding: The README instructs users to place the API token in the GUAIKEI_API_TOKEN environment variable, but it does not clearly frame the token as a sensitive secret or warn against exposing it in shell history, screenshots, shared terminals, CI logs, or committed configuration files. This omission increases the chance of accidental credential leakage and unauthorized API use.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: This documentation explicitly enables bulk collection of Douyin search results, creator posts, comments, and hot-list data at scale, with limits up to 10,000 records, but it provides no guidance on privacy, lawful basis, rate limiting, retention, or platform terms compliance. In the context of a competitive-intelligence skill, that omission increases the likelihood of misuse for mass scraping, profiling, or collection of personal data from comments and accounts.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The command silently persists fetched results to a local JSON file after execution, which can leave sensitive competitor research, scraped content metadata, or user analysis data on disk without the operator realizing it. In shared environments, CI runners, or multi-user systems, these files may be later accessed, backed up, or indexed, creating unintended data exposure.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The CLI persists full search output, including the user's keyword and returned results, to a local JSON file without an explicit warning or consent flow. In a competitive-intelligence context, these queries and results may be commercially sensitive, and silent local retention increases the risk of unintended disclosure through shared machines, backups, or later log collection.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The command sends user-supplied keywords and search parameters to remote APIs to create and retrieve search tasks, but the interface does not clearly disclose that user input is transmitted externally. Because this skill is specifically used for competitor and market-intelligence research, the search terms can reveal sensitive business interests, making undisclosed transmission a meaningful privacy and confidentiality issue.

Ssd 3

Medium

• Confidence: 95% confidence
• Finding: The skill instructs automatic archival of logs using a filename pattern containing timestamp, keyword, parameters, and type. This creates retention of user-provided search terms and operational metadata, which may include sensitive business interests, monitored accounts, or URLs, and increases exposure if the host is shared or logs are exfiltrated.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.