E-commerce Price Monitoring

Problem

E-commerce Price Monitoring targets protected data workflows in agent scenarios. It addresses two engineering issues: unreliable Python environment checks before script execution, and business reasoning that proceeds before a protected capability is authorized. It does not assume a local interpreter is available, nor does it expect callers to handle AppKey, AppSecret, JWT parsing, or binding manually.

How It Works

• Check the environment first: silently resolve a usable PYTHON_CMD; if the version is acceptable, execute directly, and report only when detection fails. On Windows, it can suggest py -3 or disabling app execution aliases.
• Enforce the authorization gate: before each use, scripts/auth.py validates local credentials, user binding, and platform checks. Signals such as AUTH_REQUIRED or CREDENTIALS_MISSING are recoverable and should trigger automatic JWT acquisition and binding, followed by a new --ensure check.
• Execute protected instructions only: the workflow may continue only after scripts/protected.py returns protected instructions, preventing business output based on local hints or stale context.

Boundaries

This is a protected capability and should be invoked after user authorization and platform permission checks are complete. If authentication fails, the network is unreachable, or remote validation fails, business output must stop. Do not retry silently, expose secrets in plaintext, or skip protected instructions to generate monitoring results.