Back to skill

Security audit

meituan-travel

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent Meituan travel assistant, but it asks users to run an unpinned npm/npx CLI that can access a sensitive Meituan token.

Review the npm package provenance before installing, prefer a pinned exact version or lockfile-based install, and only provide MEITUAN_HT_TOKEN in a trusted environment. Do not use this skill where an unreviewed future npm package version could access sensitive credentials or affect paid travel actions.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The manifest description, usage guidance, examples, and required user-facing waiting message are all written in Chinese, and the skill does not indicate that this language constraint is optional or region-specific. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding

The skill instructs users to execute an unpinned package via npx, which can fetch the latest published version at runtime. If the package is updated maliciously, compromised upstream, or a typo/package-ownership issue occurs, the skill could execute attacker-controlled code in the local environment with access to the configured token.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding

This invocation again relies on npx without an exact version pin, causing code to be downloaded and executed dynamically at runtime. In this skill, the risk is elevated because the tool is expected to access MEITUAN_HT_TOKEN, so a compromised package version could exfiltrate credentials or manipulate travel query results.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.