Security audit

Alipay

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only Alipay API skill, but it references sensitive payment, authorization, and merchant identity data that users must handle carefully.

Install only if you need Alipay API reference material. When using it, treat app authorization tokens, refresh tokens, passwords, phone numbers, business licenses, bank or settlement details, and certificate/identity data as sensitive: use sandbox or test credentials where possible, avoid logging or hardcoding secrets, redact examples before sharing, and require explicit confirmation before running mutating Alipay API calls such as create, confirm, cancel, upload, or onboarding operations.

SkillSpector

By NVIDIA

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 (6)

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The API documentation explicitly includes highly sensitive fields such as app_auth_token, app_test_account_password, and web_test_account_password, but provides no warning about secure handling, storage, masking, or use of non-production credentials. In an agent skill context, this can normalize collecting and transmitting secrets in plain examples, increasing the risk that downstream users or generated code will log, hardcode, or expose real credentials.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The documentation includes realistic-looking values for `app_auth_token`, `app_refresh_token`, `code`, and `user_id` directly in request and response examples without any clear warning that these are secrets or placeholders. In an API integration context, authorization codes and tokens are highly sensitive credentials; publishing or reusing real-looking examples can lead developers to mishandle them, accidentally expose live tokens in logs, code samples, or copied integrations.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The API doc and sample request explicitly include fields for test account credentials such as app_test_account_password and web/h5 test passwords, but provide no guidance on secure handling, masking, storage limits, or avoiding use of real credentials. In an API-integration skill, this omission increases the chance that developers transmit reusable credentials in plaintext examples, logs, tickets, or long-lived configs, leading to credential exposure.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The document requires submission of sensitive personal and business data, including business license images, contact mobile numbers, and identity-related materials, but does not mention privacy handling, data minimization, retention, consent, or secure storage expectations. In onboarding/payment contexts, this can lead integrators to over-collect, improperly retain, or insecurely process regulated merchant data.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The documentation enumerates highly sensitive merchant and personal identity fields such as phone numbers, login IDs, certificate numbers, legal-person identity data, and beneficiary document details, but provides no warning about secure handling, minimization, storage restrictions, or masking. In an agent skill context, this increases the chance that downstream tooling or users will copy, log, retain, or test with real PII and account data without appropriate safeguards.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The request example includes realistic sensitive identity and account data patterns, including phone numbers, login IDs, certificate numbers, names, addresses, and document image references, without any warning that examples must be fictitious, redacted, or non-production. This is dangerous because users may reuse the sample structure with live data, and agent systems may capture or surface these values in prompts, logs, traces, or debugging outputs.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.