Security audit

抖音每日热门作品榜

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Douyin ranking lookup tool that uses a RedFox API key and fixed RedFox endpoint, with only minor scoping and documentation caveats.

Before installing, confirm you trust RedFox with REDFOX_API_KEY and with the Douyin ranking queries you make. Treat subscription claims as documentation only unless the host provides a separate subscription mechanism, and prefer explicit Douyin-ranking prompts to avoid accidental use.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (6)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 92% confidence
Finding: The skill instructs use of an environment-sourced API key and outbound network access, but no corresponding permissions are declared. This creates a transparency and policy-enforcement gap: a host may permit the skill to run without clearly signaling that it can access secrets and call external services, increasing the risk of unintended credential use or data exfiltration through the documented API workflow.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 84% confidence
Finding: The documented behavior does not fully match the implemented capabilities and constraints: the skill appears to support additional query parameters and output sizing while failing to enforce advertised limits like 30-day history and subscription behavior. These mismatches can lead to over-broad data access, operator misunderstanding, and unsafe reliance on controls that users or reviewers believe exist but are not actually enforced.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The README explicitly instructs users to invoke the skill with unconstrained natural language ('Simply describe your needs in natural language'), which increases the chance the skill is selected for loosely related prompts rather than clearly scoped requests. In an agentic environment, broad invocation guidance can cause unintended tool use, unnecessary external API calls, and disclosure of user intent to a third-party service even when the user did not clearly request this specific skill.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The example phrases include very generic language such as 'Today's ranking' and 'Show all,' which are common across many contexts and can make the orchestrator route unrelated requests to this skill. That broad matching is risky because it may trigger external requests and expose user queries to the RedFox-backed service without sufficiently specific user intent tied to Douyin content.

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The README states users can 'directly use natural language to describe needs' without defining clear invocation boundaries. In an agent ecosystem, this broad trigger guidance can cause the skill to activate on loosely related requests, increasing the chance of accidental invocation, context hijacking, or routing away from more appropriate tools.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: Several example phrases are short and generic, such as requests equivalent to 'today's ranking' or 'view all,' which may overlap with ordinary conversation or other ranking tasks. This raises the risk that the skill is invoked unintentionally, potentially exposing external API usage, irrelevant results, or tool-selection confusion in multi-skill environments.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.