ClawHub
Back to skill

Security audit

美团优惠券自动领取

Security checks for vulnerabilities and agentic risk

Overview

This skill does what it claims, but it can automate actions on a logged-in Meituan account and set recurring scheduled runs without enough consent and disablement guidance.

Install only if you are comfortable with the agent using your logged-in Meituan browser session to click coupon-claim controls. Before enabling the daily schedule, confirm exactly when it will run and how to list, pause, or delete the scheduled task.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger phrase "查看优惠券" is broad enough to match ordinary user requests that may simply ask to view coupon information, but this skill performs browser automation and coupon-claiming actions. That creates a risk of unintended invocation and unauthorized state-changing actions on the user's Meituan account.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill includes instructions to create a scheduled task that will repeatedly perform browser-based coupon collection, but it does not prominently warn the user that this is recurring autonomous behavior on their behalf. Persistent automation against an authenticated account can surprise users, create unintended repeated actions, and make consent and revocation unclear.

Vague Triggers

Low
Confidence
91% confidence
Finding
The listed trigger phrases define a wide invocation scope but do not state exclusions, boundaries, or examples of when the skill should not run. Because the skill opens a browser, clicks through a third-party page, and can set recurring automation, ambiguous triggering increases the chance that normal conversation will launch account-affecting actions without clear consent.

Static analysis

No suspicious patterns detected.