Security audit

小红书视频下载

Security checks across malware telemetry and agentic risk

Overview

This skill does what it says: it sends Xiaohongshu video links to redfox.hk to return downloadable video links, with some privacy and lawful-use caveats users should understand.

Install only if you are comfortable sending Xiaohongshu links and your redfox.hk API key to redfox.hk. Prefer the REDFOX_API_KEY environment variable, avoid --save-key or manual plaintext config storage on shared/synced machines, and only download or reuse videos when you have the right to do so and can comply with platform terms and local law.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (8)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 92% confidence
Finding: The skill advertises simple link parsing, but the analyzed behavior indicates additional sensitive operations: reading/storing API keys locally, batch processing, collecting richer metadata, and promotional side behavior. This mismatch is dangerous because users may provide credentials and links without informed consent, increasing the risk of credential exposure, overcollection, or unexpected data handling.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The skill prominently advertises downloading third-party videos and removing watermarks, but provides no warning about copyright, consent, platform terms, or privacy implications. In this context, the omission increases the likelihood of misuse for unauthorized copying, redistribution, or harvesting of other users' content, especially because the skill is explicitly framed for collection, backup, and remixing.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The skill explicitly promotes downloading watermark-free third-party videos, backup of others' content, and reuse for secondary creation without any warning about copyright, authorization, or platform-policy limits. In context, this increases the likelihood of misuse for unauthorized copying and redistribution, making the skill materially riskier than a neutral link parser.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill asks users to paste Xiaohongshu links and states it uses redfox.hk for parsing, but it does not clearly warn that submitted links are sent to a third-party service. Even if the links are not credentials, they may reveal user interests, private/unpublished content references, or business research activity, creating a privacy and data-sharing risk.

Missing User Warnings

Low

Confidence: 88% confidence
Finding: The instructions tell users to place an API key in an environment variable without any warning about secret handling, rotation, or avoiding logs/shell history. This can lead to accidental credential disclosure, especially in shared machines, copied terminal sessions, or misconfigured tooling that prints environment variables.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The workflow instructs immediate processing of user-supplied Xiaohongshu links, but the skill documentation later shows that parsing is performed by sending the URL to the external service redfox.hk. That creates a privacy and data-sharing risk because users are not informed that their provided link will be transmitted to a third party, which may expose personal, private, or tracking-bearing URLs.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The documentation offers a --save-key option that persists the API key to a local configuration file, but it does not warn users that this stores credentials on disk. Storing secrets locally without explaining filesystem exposure, backup leakage, or shared-account risks can lead to credential compromise and unauthorized API use.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The example command writes the API key directly into ~/.qoder/apis/redfox.json, normalizing plaintext credential storage without any caution. This increases the chance that users expose the key through permissive file permissions, shell history, backups, synced home directories, or other local disclosure paths.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.