earnings-calendar

name	earnings-calendar
description	Use when the user asks "any earnings coming up?", "who reports this week?", "earnings calendar", "what's reporting?", or any variation about scheduled earnings events. Also use proactively during daily-update to flag tickers reporting within the next 7 days, and when presenting a signal on a stock that reports soon.

Earnings Calendar

Earnings reports are scheduled E[CF] events. Unlike news (which you react to after it happens), earnings dates are known in advance. Earnings are the single largest E[CF] catalyst type. A BUY signal 2 days before earnings carries different risk than one 2 weeks after. This skill surfaces upcoming earnings so you can factor scheduled catalysts into signal interpretation and position sizing decisions.

The script

scripts/earnings.py scans all tracked tickers and reports upcoming earnings dates. Results are cached at data/_meta/_earnings.json with event-based staleness: a cached date is reused as long as it's in the future. Once the earnings date passes, the next quarter's date is fetched automatically. First run fetches all tickers; subsequent runs only refetch tickers whose dates have passed.

./run .claude/skills/earnings-calendar/scripts/earnings.py ~/.lafmm/data
./run .claude/skills/earnings-calendar/scripts/earnings.py ~/.lafmm/data --group semis --days 7
./run .claude/skills/earnings-calendar/scripts/earnings.py ~/.lafmm/data --cal
./run .claude/skills/earnings-calendar/scripts/earnings.py ~/.lafmm/data --json

Options:

• --group NAME — scan only this group (directory name, e.g. semis).
• --days N — lookahead window in calendar days. Default: 14.
• --cal — display as a colored month grid instead of a list.
• --json — machine-readable output.

Interpreting the output

The script lists tickers with earnings within the window, sorted by date. For each ticker it shows the group, the date, and how many days away it is.

What matters is not the earnings date itself but what it means for your current positions and signals:

• Signal + earnings in 1-3 days: the signal is based on trend. Earnings can override the trend in either direction. The user decides whether to hold through or reduce exposure before the event.
• Signal + earnings just passed: post-earnings price action is often the most reliable. The market has repriced E[CF]. A BUY signal that forms after an earnings beat is stronger than one before.
• No signal + earnings soon: be alert. Earnings can trigger the column transition that produces a new signal. Watch for WATCH alerts (pivot proximity) around the earnings date.

Requirements

The skill requires tracked tickers with price data in the data directory. Internet access is required for the initial fetch and for refreshing dates after earnings pass.

When to use this

• When the user explicitly asks about upcoming earnings dates.
• Proactively during daily-update (step 4) to flag tickers reporting within 7 days.
• When presenting any signal, to check if earnings are imminent.
• Before sizing a new position, to understand catalyst timing.
• When a ticker shows unusual volume or price action without obvious news.

Limitations

Yahoo Finance earnings dates are best-effort. Dates can shift, some tickers return no data, and the timing field (pre-market vs after-hours) is not always available. Treat this as a heads-up, not an authoritative source. For exact timing, the user should verify with their broker or the company's investor relations page.