抖音视频下载器

• Outcome: Review
• Latest audit: Jul 28, 2026 · 12:03 PM UTC
• Version: 1.0.3

Overview

The skill is a disclosed video downloader that sends supported media links to redfox.hk and saves returned media locally, with no hidden persistence or unrelated access found.

Install only if you are comfortable sending submitted media links to redfox.hk and saving the resulting files locally. Use it only for content you have rights to download, avoid private or tokenized links, and prefer environment-variable or protected config-file API key storage over command-line keys.

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: 93% confidence
• Finding: The skill documentation advertises and invokes a Python downloader that uses environment variables, local file reads/writes, network access, and shell execution patterns, but the skill declares no permissions. This creates a transparency and governance gap: users and hosting platforms cannot accurately assess what the skill will access, and may run it with broader trust than warranted.

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The invocation guidance tells users to 'just describe what you want in natural language' without clearly constraining allowed actions, scope, or confirmation requirements. In an agent setting, this broad trigger language can cause the skill to activate on loosely related prompts and perform network/download actions on untrusted URLs without sufficient user awareness, increasing the chance of unintended file writes or misuse.

Missing User Warnings

Medium

• Confidence: 81% confidence
• Finding: The README promotes downloading and permanently saving third-party content locally, including watermark-free media, but does not warn about privacy, copyright, storage, or the risks of downloading content from external platforms. In an agent environment, this omission can lead users to trigger network retrieval and local file persistence without informed consent about data handling or system impact.

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The README tells users to invoke the skill with unrestricted natural-language requests such as downloading arbitrary links, without defining clear boundaries, consent checks, or content-safety constraints. In a downloader skill that fetches and saves third-party media locally, this increases the chance of misuse for unauthorized content acquisition, accidental processing of unsafe URLs, or abuse beyond the intended scope.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The usage section encourages downloading and saving external content locally but does not prominently warn users about the privacy, copyright, malware, and storage risks of fetching third-party media. Because the skill specifically targets multiple external platforms and stores files on disk, omission of these warnings makes unsafe or non-compliant use more likely.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill sends user-supplied share links to the third-party service redfox.hk for parsing and download, but the description does not clearly warn users that their input is transmitted off-platform. Shared links can contain personal, private, or access-scoped content, so undisclosed third-party transmission creates privacy, consent, and compliance risk.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The examples show passing the API key directly on the command line, which can expose credentials through shell history, process listings, logs, or screenshots. Presenting this without a warning normalizes an unsafe handling pattern and increases the chance of accidental credential leakage.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The tool sends the user-supplied video URL to redfox.hk for parsing without an explicit consent prompt or prominent disclosure at the point of transmission. Shared URLs can contain private identifiers, tracking parameters, or access tokens, so silent disclosure to a third party creates a real privacy and data-handling risk.

Missing User Warnings

Medium

• Confidence: 78% confidence
• Finding: The presentation encourages downloading third-party short-video content for local saving without any warning about copyright, authorization, or local file effects. In this skill context, that omission increases legal/compliance risk and may normalize unsafe or unauthorized content acquisition workflows for end users.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The configuration content discusses API keys and a third-party API but does not warn users that pasted share links and related request data will be transmitted to an external service. In a downloader skill, this matters because shared links can reveal user interests, private or semi-private content references, and account-associated activity to an outside provider.

Missing User Warnings

Medium

• Confidence: 84% confidence
• Finding: Promoting downloaded content for remixing or secondary creation without a rights warning can facilitate misuse of copyrighted material. In this skill's context, the feature is directly tied to obtaining third-party media, so omission of permission and licensing guidance raises compliance and abuse risk.

VirusTotal

65/65 vendors flagged this skill as clean.

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical

• Code: suspicious.exposed_secret_literal
• Location: assets/downloader.py:30