Back to skill

Security audit

obtain-coupons-all-in-one

Security checks for vulnerabilities and agentic risk

Overview

The skill generally fetches coupons as described, but it creates and sends a persistent install identifier for attribution and presents broad third-party coupon flows with limited user control or privacy detail.

Review this skill before installing if you are privacy-sensitive. It contacts agskills.moontai.top and returns third-party coupon links or QR codes, and the bundled script stores a local install ID that is sent for attribution on requests. Install only if you are comfortable with China-market coupon and WeChat-style redemption flows and with that persistent attribution behavior.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (15)

Rp1

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The README instructs users to install and execute a remote package via npx skills add moooai/obtain-coupons-all-in-one without pinning a specific version or immutable reference. That exposes users to supply-chain risk because a later compromised or maliciously updated package could be fetched and run implicitly, and the README provides no integrity verification guidance.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The suggested invocation phrases are very broad, natural-language requests such as '帮我领今天所有的优惠券' and '外卖和快递的优惠一起看看', which can overlap with ordinary user intent and cause the skill to trigger unexpectedly. In an agent ecosystem, broad triggers increase the chance of unintended activation, especially because this skill can initiate external data fetching and coupon collection actions across multiple platforms.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The README states that API requests automatically append ch and u attribution parameters tied to an anonymous installation identifier, but it does not clearly explain what data is collected, how long it is retained, who receives it, or how users can opt out. This creates a privacy and transparency issue because users may unknowingly transmit persistent tracking metadata whenever the skill is used.

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill instructs the agent to call multiple third-party APIs and distribute returned links/QR codes, but it does not clearly warn users that their requests may be sent to external services or explain the privacy implications. Even if no authentication is required, user queries, IP metadata, device/network context, and downstream tracking parameters may be exposed to outside parties without informed consent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The file repeatedly instructs users to open links or scan codes in WeChat and presents all usage examples and operational guidance in Chinese-centric workflows, but it does not offer alternatives or state that the skill is intended only for a China-specific audience. This can violate language/locale policy because it imposes a specific locale-dependent experience without explicit opt-in or justification.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases are extremely broad, such as generic questions about discounts or coupons, which can cause the skill to activate during ordinary conversation and unexpectedly steer users into external coupon flows. This increases the chance of unwanted network access, confusing behavior, and unsolicited presentation of third-party links or QR codes.

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

SQP-3 covers natural-language policy violations such as forcing a specific language or locale without user opt-in. This file presents all instructions in Chinese and recommends a China-specific app ecosystem, but does not indicate that the skill is region-specific or provide any language choice.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The documentation directs users to open third-party coupon URLs and scan QR codes without warning that those actions may disclose identifiers, referral parameters, device metadata, or behavioral tracking data to external services. Because the links contain apparent affiliate or account identifiers, users may be silently funneled through tracking infrastructure or promotional attribution flows without informed consent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The module docstring and all user-facing text are exclusively in Chinese, indicating the skill is designed around a fixed language/locale without any opt-in or alternative. The policy allows locale constraints only when choice is offered or the restriction is clearly justified, which is not present here.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The script generates a persistent install identifier, stores it locally, and automatically transmits it as the 'u' query parameter to a remote coupon service for attribution/tracking. The docstring frames this as a first-run local persistence detail, which obscures the fact that the identifier enables cross-run correlation of a user's activity by the remote operator without any user-facing disclosure or consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The code writes a stable UUID to scripts/.install_id and reuses it on every request, sending it to an external domain together with a channel tag. This creates a durable tracking token that can link a user's requests over time, and because there is no warning, consent flow, or easy opt-out, it is a privacy-impacting behavior inconsistent with user expectations for a simple coupon-fetching tool.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The skill metadata presents its user-facing description entirely in Chinese, and the manifest does not indicate that this is a China-specific or Chinese-only skill, nor does it offer a language choice. Under the stated policy, a locale or language constraint should be documented or optional rather than implicitly imposed.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The example code performs direct requests to third-party endpoints without any warning that use of the examples will generate outbound network traffic to non-local services. This can surprise users in restricted or privacy-sensitive environments and may expose IP addresses, user agents, or operational metadata to external infrastructure.

Content

No source excerpt is available for this finding.