TikTok官方-带货视频API

• Outcome: Review
• Latest audit: Aug 14, 2026 · 3:49 PM UTC
• Version: 1.0.4

Overview

This is mostly a coherent TikTok video API helper, but it needs Review because it handles creator tokens and public posting while also instructing automatic feedback reporting and sensitive response persistence.

Install only if you are comfortable giving the skill access to LinkFox/TikTok creator tokens and letting it perform user-directed TikTok video operations, including public posting. Avoid passing raw tokens on the command line when possible, do not store response dumps in shared or synced directories, delete persisted API responses after use, and disable or require explicit approval for any Feedback API submission.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection

Findings (7)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 84% confidence
• Finding: The skill exposes effective capabilities for shell, network, file read/write, and environment access, but does not declare permissions or place visible constraints around their use. In practice this reduces reviewability and makes it easier for the skill to perform sensitive actions such as storing API responses, invoking external services, or handling tokens in ways the caller may not expect.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 92% confidence
• Finding: The skill's declared purpose says it handles TikTok video upload and large-file upload flows, but the implementation only provides guidance for binary/chunk upload while also bundling an undeclared response persistence utility. That mismatch can mislead an agent into invoking the skill in contexts where it cannot safely complete the task, while still enabling unrelated sensitive behavior such as writing full API responses containing PII or tokens to disk.

Context-Inappropriate Capability

Medium

• Confidence: 90% confidence
• Finding: The skill instructs the agent to automatically call a separate Feedback API whenever certain conditions occur, even though that action is unrelated to the user's requested TikTok video operation. This creates an unnecessary outbound side effect and risks exfiltrating user content, operational metadata, or sensitive context to another service without explicit user awareness or consent.

Vague Triggers

High

• Confidence: 88% confidence
• Finding: The trigger description is extremely broad, including many loosely related phrases and instructions to activate even when authorization is not explicitly mentioned. Over-broad invocation increases the chance this skill is selected in unintended contexts, which is risky because the skill can reach networked APIs, handle access tokens, and persist responses to disk.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill explicitly recommends persisting full API responses to disk and notes those responses may contain PII or auth-sensitive data, but this behavior is not surfaced as a major security-sensitive feature in the main behavior description. Writing sensitive API data to local storage creates confidentiality and retention risks, especially if stored in insecure temp locations, left behind after execution, or read by other processes/users.

Missing User Warnings

Low

• Confidence: 87% confidence
• Finding: The script’s usage examples explicitly encourage passing a TikTok access token as a command-line argument. On many systems, command-line arguments are exposed via shell history, process listings, audit logs, or orchestration tooling, which can unintentionally leak credentials; in this skill context, the token is then used to access creator/video APIs, making accidental disclosure meaningful.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The helper always writes captured stdout to disk, including potentially sensitive API responses, tokens, creator/shop data, or business records, without any consent prompt, warning, retention control, or redaction. In this TikTok upload skill context, persisted responses may contain account metadata and operational data that become exposed to other local users, logs, backups, or later processing.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.