Security audit

guaikei-xhs-fetcher

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Xiaohongshu public-data fetcher that uses a guaikei API token and saves successful results locally.

Install only if you are comfortable sending Xiaohongshu keywords, links, and API token-authenticated requests to guaikei.com. Review or delete the local logs directory after use if fetched comments, URLs, or research terms are sensitive.

SkillSpector

By NVIDIA

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 (5)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 91% confidence
Finding: The skill requires access to a sensitive environment variable (`GUAIKEI_API_TOKEN`) but does not declare corresponding permissions or clearly constrain how that secret is handled. This creates a governance gap: the runtime can use external credentials and call a third-party service without an explicit permission boundary, making secret exposure or unauthorized outbound use harder to audit and control.

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The trigger conditions are broad enough that the skill may activate on loosely related requests, including generic analysis or content tasks that merely mention xhs/rednote. Over-broad routing increases the chance of unnecessary third-party data transmission and execution when the user did not explicitly consent to using this external integration.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: Although the document mentions a third-party API later, the main capability description emphasizes functionality without prominently warning that user-provided keywords, URLs, and query parameters will be sent to `guaikei.com`. This can undermine informed consent and lead to unintended disclosure of user-supplied research targets, links, or monitoring interests to an external service.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The CLI persists full comment task output to a local JSON file, which may include scraped comment content and associated metadata, without clear user notice or any retention controls in this file. On shared machines or CI/runtime environments, this can unintentionally expose collected data to other local users, backups, or log collection systems.

Missing User Warnings

Medium

Confidence: 82% confidence
Finding: Search results are persisted to a local file without an explicit nearby warning, consent, or an opt-in control. In this skill context, returned data may include note content, comments, and keyword-derived research artifacts that can be sensitive from a privacy/compliance perspective, so silent disk persistence increases exposure if the host is shared, monitored, or later compromised.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: SKILL.md:16