Shipment Tracker

Track packages across multiple carriers from a markdown shipments file. Auto-detects carrier from tracking number patterns. Hybrid status checking: tries direct HTTP first, falls back to recommending browser-use for full tracking details.

Shipments File Format

The skill reads a markdown file with a table of active shipments:

# Active Shipments

| Order | Item | Carrier | Tracking | Link | Added |
|-------|------|---------|----------|------|-------|
| Acme #1234 | Widget | USPS | 9449050899562006763949 | [Track](https://...) | 2026-02-19 |

• Carrier and Link are optional — auto-detected from tracking number pattern
• Remove entries once delivered to keep the file clean
• Default location: memory/shipments.md in the workspace

Usage

# Check all active shipments
python3 scripts/shipment_tracker.py memory/shipments.md

# JSON output for integrations
python3 scripts/shipment_tracker.py memory/shipments.md --format json

# Detect carrier from a tracking number
python3 scripts/shipment_tracker.py --detect 9449050899562006763949

Carrier Detection

Automatically identifies carrier from tracking number patterns:

Status Checking (Hybrid)

• Direct HTTP — Attempts to extract status from carrier tracking pages via urllib. Works for basic status on USPS and some other carriers.
• Browser-use fallback — When HTTP fails or carriers use JS-heavy pages, the script provides the exact browser-use command to run.

python3 -c "
import asyncio
from browser_use import Agent, Browser, ChatBrowserUse
async def main():
    browser = Browser(use_cloud=True)
    llm = ChatBrowserUse()
    agent = Agent(
        task='Go to <tracking_url> and extract the current tracking status, delivery date, and location',
        llm=llm, browser=browser
    )
    result = await agent.run(max_steps=10)
    print('TRACKING:', result)
asyncio.run(main())
"

This ensures reliable tracking across all carriers, even those with aggressive bot detection.

When browser-use is needed:

• UPS, FedEx, Amazon (heavily JS-based tracking pages)
• USPS when basic parsing fails (complex status updates)
• Any carrier with CAPTCHA or bot detection
• Sites that require user interaction or form submission

Workflow

• User provides a tracking number → run --detect to identify carrier
• Add to memory/shipments.md with order details
• Morning briefing or on-demand: run the script to check all shipments
• For shipments needing browser-use:
• Non-sensitive packages: Use the provided browser-use command
• Privacy-sensitive packages: Manual browser check instead (data stays local)
• When delivered: remove from the shipments file

System Access

Direct skill execution:

• File reads: One markdown file (path provided as argument)
• Network: HTTPS GET to carrier tracking pages (e.g., tools.usps.com) — read-only, no authentication
• No file writes, no subprocess calls, no shell execution

• Tracking numbers and order information sent to cloud browser service
• Package details processed by external LLM (ChatBrowserUse)
• Carrier tracking URLs accessed via cloud infrastructure

Requirements

• Python 3.10+
• Outbound HTTPS access to carrier tracking sites
• browser-use (optional, for full tracking on JS-heavy sites)