skills/alipay/ai/alipay-aipay/Gen Agent Trust Hub

alipay-aipay

Pass

Audited by Gen Agent Trust Hub on Nov 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads several legitimate dependencies from trusted or well-known sources during its initialization and execution phases:
  • Fetches the alipay-cli tool from the official Alipay domain (opengw.alipay.com).
  • Downloads jq from its official GitHub repository (jqlang/jq).
  • Bootstraps a compatible Node.js environment for Windows from the official Node.js distribution site (nodejs.org).
  • [REMOTE_CODE_EXECUTION]: The skill uses piped shell execution (curl | bash) to install the alipay-cli tool from the vendor's official domain. It also uses npx to perform self-updates by downloading and executing the latest version of the skill's own package (@alipay/alipay-aipay) from the public npm registry.
  • [COMMAND_EXECUTION]: The skill executes various system commands through its management scripts to perform environment checks, manage local configuration files, and facilitate sandbox testing. For example:
  • The alipay_cli_refresh.mjs script manages the installation and version verification of the Alipay CLI.
  • The local_402_sandbox_pay.py script uses subprocess.run to interact with local services and the sandbox payment gateway via curl.
  • [DATA_EXFILTRATION]: The skill includes an opt-in telemetry and conversation collection feature.
  • Usage metrics and stage completion events are reported to the vendor's telemetry service (collect.alipay.com).
  • Conversation logs can be collected for product improvement, but this requires the user to provide explicit consent via the conversation consent tool.
  • The collection logic in conversation_telemetry.mjs contains extensive redaction patterns to mask private keys, passwords, and personally identifiable information (PII) before transmission.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 19, 2026, 12:34 PM