Stock Monitor

Track product availability on Shopify stores. Get notified via SMS, Slack, or terminal when out-of-stock items come back.

Quick Start

# Add a product to watch (auto-detects name from Shopify)
python3 ~/.claude/skills/stock-monitor/scripts/stock_watch.py add "https://store.com/products/item?variant=123" --notify sms --target "+17327595647"

# Check all watched items
python3 ~/.claude/skills/stock-monitor/scripts/stock_check.py

# One-off check (no watchlist needed)
python3 ~/.claude/skills/stock-monitor/scripts/stock_check.py --url "https://store.com/products/item?variant=123"

Scripts

stock_watch.py — Manage Watchlist

Command	Purpose
stock_watch.py add URL [--name NAME] [--notify sms|slack|terminal] [--target PHONE|#CHANNEL]	Add product to watch
stock_watch.py list	Show all watched items with status
stock_watch.py remove ID	Remove a watch
stock_watch.py pause ID / resume ID	Pause or resume checking

stock_check.py — Run Checks

Flag	Purpose
(no flags)	Check all active watches
--id N	Check specific watch
--url URL	One-off URL check, no DB write
--dry-run	Check without sending notifications
--json	JSON output for piping

Notification Channels

• terminal (default): Colored output to stdout
• sms: Dispatches via ~/.claude/skills/sms/scripts/sms_send.py
• slack: Dispatches via ~/.claude/skills/slack/scripts/slack_post.py

Notifications fire only on out_of_stock → in_stock transitions to avoid noise on repeated checks.

Scheduling

Run stock_check.py manually, or schedule recurring checks:

• Session loop: /loop 30m python3 ~/.claude/skills/stock-monitor/scripts/stock_check.py
• Persistent cron: See references/launchd-setup.md for a macOS launchd plist that runs checks when Claude is closed.

How It Works

Shopify stores expose product data at /products/{handle}.js with per-variant available booleans. The skill fetches this JSON, finds the target variant by ID, and compares against the last known status in SQLite. No scraping, no browser rendering, no rate-limit concerns.

Data

SQLite DB at ~/.claude/skills/stock-monitor/data/stock_monitor.db with two tables: watches (watchlist) and check_log (check history with transition tracking).