Security audit

guaikei-douyin-data-feed

Security checks for vulnerabilities and agentic risk

Overview

This skill mostly does what it says, but its broad triggers, automatic data logging, and runtime token-error messages need human review before installation.

Review before installing. Use this only when you intend to send Douyin queries, URLs, and requested public result data to Guaikei's API and keep local JSON logs. Store GUAIKEI_API_TOKEN as a secret, avoid shared machines or synced folders for sensitive searches, periodically delete logs you no longer need, and confirm ambiguous user requests before invoking the skill.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (7)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 90% confidence
Finding: The skill explicitly relies on the environment variable `GUAIKEI_API_TOKEN`, but there is no declared permission boundary corresponding to environment access. This creates a mismatch between documented runtime behavior and the skill's permission model, making secret access less visible to reviewers and increasing the chance of over-broad or unexpected credential exposure.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The skill says it should be used 'even if the user did not mention Douyin,' which broadens routing beyond clear user intent. That can cause the agent to invoke an external data-access skill for ambiguous requests, leading to unintended data retrieval, incorrect platform assumptions, and reduced user control over when third-party/API-backed actions occur.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The README encourages collecting public videos, account data, comments, and exporting them to local logs, but does not clearly warn users about privacy, retention, downstream sharing, or compliance obligations. Even when data is public, bulk collection and local storage of comments and profile-related data can increase privacy and misuse risks, especially in marketing, competitor monitoring, and sentiment analysis contexts.

Missing User Warnings

Low

Confidence: 82% confidence
Finding: The documentation tells users to configure GUAIKEI_API_TOKEN but does not prominently state that it is a sensitive credential used for outbound API requests and must not be exposed in logs, screenshots, or shared environments. This increases the chance of accidental credential leakage, which could allow unauthorized API use or quota abuse.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The CLI sends the user-supplied Douyin URL to a remote API to retrieve comments and later writes the returned comment data to a local JSON file, but the code provides no explicit consent prompt, privacy notice, or option to disable persistence. This creates a privacy and data-governance risk because users may not realize that queried content and fetched comments are transmitted off-host and retained on disk, which can expose sensitive browsing targets or collected data to other local users, backups, or logging systems.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The CLI persists the full retrieved Douyin results to a local JSON file automatically, without an explicit opt-in or warning at the point of use. Because the results may include scraped content, profile data, or other potentially sensitive information, this behavior can create unintended data retention on disk and expose the data to other local users, backup systems, or later misuse.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The script persists full search output to a local file named with the user's keyword, without clear user disclosure or consent. This can leak potentially sensitive search terms and result data to other local users, backups, telemetry collectors, or shared workspaces, especially because the filename itself embeds the query.

Static analysis

No suspicious patterns detected.