AI Agent Hub
Back to skills
Stock Guardian icon

Stock Guardian

Professional Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md to install @user_ef64daef/gushi.

About this skill

What It Solves

Staring at a watchlist is usually not the hard part. The hard part is missing the event that matters: a threshold breach, a stop target, pre-market movement, or an end-of-day summary. stock-guardian is a price watcher, not a trading advisor. You define watched tickers, thresholds, target prices, and notification channels in config.json, and the platform schedules checks to alert you when conditions trigger.

How It Works

The core loop is: read config.json, check trading hours, fetch quotes, compute changes, compare threshold/target levels, apply cooldown and alert tier, aggregate non-urgent notifications, then write history.jsonl. It adapts to the AI platform: realtime uses WebSocket, high-freq uses fine-grained cron, and patrol uses heartbeat/schedule fallbacks. Quote sources prefer Sina Finance, Tencent Finance, and Yahoo Finance APIs, with search fallback when APIs are unavailable.

Boundaries

It focuses on monitoring, alerts, candlestick observation, and backtesting. It does not provide stock recommendations, buy/sell signals, or automated trading. Free quote feeds can be delayed, and stop alerts should not be treated as the only risk control. Urgent stop alerts bypass cooldown and are sent separately, while normal alerts are aggregated in a default 5-minute window to reduce noise.

Use Cases

  • Set rise/fall thresholds for multiple A-share tickers and receive Feishu or email alerts during market hours without continuous watch while working.
  • Configure target and stop prices for US or HK stocks, and get tiered notifications when take-profit or stop levels are reached.
  • Generate a post-market summary at 15:10 with closing changes, triggered alerts, and next-day items for watched stocks automatically during monitoring.
  • Backtest historical quote data to review threshold trigger frequency, K-line pattern win rates, and whether to widen or tighten alert parameters.

Best For

  • Individual investors who track A-share, HK, and US tickers without manually polling quotes.
  • Active watchers who want threshold alerts routed to Feishu, email, or WeChat with urgent stop alerts sent separately.
  • Quant enthusiasts who want to backtest threshold frequency and K-line pattern performance before adjusting parameters.
  • Side monitoring operators who need pre-market anomalies, post-market daily reports, and weekly summaries.