Daolv Hotel Search

Security checks across malware telemetry and agentic risk

Overview

This is mainly a hotel search helper, but it should be reviewed because it includes a shared bearer token and inconsistent booking-related documentation.

Install only if you are comfortable sending hotel search details to the ai-go-hotel MCP service and using the bundled credential. Prefer a revised version that replaces the embedded bearer token with a user-supplied scoped API key and updates all documentation to consistently describe daolv-hotel-search as search/shortlist only.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The distribution instructions reference and publish a different skill slug/path (`daolv-hotel-booking`) than the actual skill under review (`daolv-hotel-search`). This can cause operators to package or publish the wrong artifact, creating a supply-chain integrity issue where users receive a different skill than intended or updates are applied to the wrong package.

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The promotional copy materially overstates the skill’s capabilities by advertising booking, room-type comparison, and pre-order confirmation even though the declared scope is search-only. This can mislead users or downstream agents into relying on unsupported actions, causing unsafe automation decisions, incorrect expectations, and possible misuse of connected MCP tools.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal