Security audit

guaikei-xhs-insight

Security checks across malware telemetry and agentic risk

Overview

This skill coherently fetches public Xiaohongshu data through a disclosed third-party API and saves results locally as documented.

Install only if you are comfortable sending Xiaohongshu keywords or URLs to guaikei.com with your GUAIKEI_API_TOKEN and having returned datasets saved under the skill's logs directory. Treat those logs as potentially sensitive business research and delete or protect them when needed.

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 (7)

Description-Behavior Mismatch

Medium

Confidence: 93% confidence
Finding: The CLI persists fetched comment results to a local JSON file via log.taskWrite after already printing the full structured result to stdout. Because comment data may contain personal or sensitive user-generated content, this creates an additional undisclosed storage surface that can expose data to other local users, backups, or later compromise, and it exceeds the stated behavior of simply returning structured data.

Description-Behavior Mismatch

Low

Confidence: 95% confidence
Finding: The CLI writes the full fetched note detail/comment payload to a local JSON file after printing results, which creates an undocumented persistence channel for potentially sensitive scraped content. Even if the data is publicly accessible, comments, identifiers, and derived metadata may be retained longer than users expect and could be exposed to other local users, backups, or log collectors.

Description-Behavior Mismatch

Medium

Confidence: 89% confidence
Finding: The CLI persists the full fetched result set to a local JSON file after printing it, but the help text and skill description only describe returning structured public data. Even if the source data is public, automatic local persistence can unexpectedly retain scraped datasets, create privacy/compliance issues, and expose data to other local users, backups, or downstream tooling without the operator's awareness.

Missing User Warnings

Medium

Confidence: 78% confidence
Finding: Automatically saving all task results to local files creates unintended data persistence, which can expose scraped datasets, searched keywords, URLs, and analysis targets to other local users, backup systems, or later compromise of the host. In this skill context, the tool is used for competitive monitoring and comment analysis, so retained artifacts may reveal sensitive business interests even if the source data is public.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The command sends the supplied note URL and retrieval request to a remote API using GUAIKEI_API_TOKEN, then writes the returned comments to disk, but the user-facing flow does not clearly warn that data will be transmitted to an external service and persisted locally. In a data-collection skill, this transparency gap can lead to unintended handling of scraped content and metadata, increasing privacy, compliance, and operational risk.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: Task results are written to a local JSON file without explicit user consent in the execution path, creating silent retention of retrieved note/comment data. This is more concerning than ordinary console output because the file may persist sensitive or regulated content, survive session boundaries, and be accessible to other processes or users on the host.

Missing User Warnings

Low

Confidence: 85% confidence
Finding: The tool writes result data to a local JSON file without clearly informing the user in help text or requiring consent. This weakens operator awareness and can cause unintended retention of collected data on disk, especially in shared environments or automated pipelines.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.