What is coming up, and what does it mean for the book?

Paste the upcoming events for your coverage list and your positions. Your browser cleans the list as you type - dates normalised, duplicates dropped, each event typed and given an H/M/L impact, grouped by week, conflicts and unconfirmed dates flagged - and exports it to your calendar, free, before you sign in. Then the weekly preview says what matters this week, what to prepare for next week, and what to do before each event.

Each example comes with a saved model run, so you can see the whole page for free.

As of

Horizon

Next week's run

Upcoming events

A table with a Date column (pipes, tabs or CSV; optional Time, Ticker, Event, Type, Impact, Notes), or one event per line starting with a date. Mark unconfirmed dates "(est)"; write TBD, "week of", "early Nov" or "Q4 2026" when there is no exact day. Use MACRO as the ticker for market-wide events.

Your positions (optional, but it is what makes the preview yours)

Drop a .csv, .tsv or .txt event list, or a set .json saved from this page, or

What you want to know (optional)

Paste the events to price the preview.

this week

high-impact covered

positions addressed

flags answered

statements checked

disagreements

This week's key events

Next week

Position implications

Position	Events	Exposure	Implication	Before the event	Risk

Dates to verify

Gaps in the list

Impact changes

Browser flags, answered

Reconciliation with your calendar

Every event, position and flag id had to exist, every event under this week or next week had to be in that window, every impact had to be the calendar's (or an explicit change), every high-impact event this week and every position with a high-impact or binary event had to be covered, "binary" had to rest on a binary event, every flag had to be answered, and every $ or % figure and every ticker in the prose had to be in your paste.

Weekly preview note

Raw model reply

The reply did not parse as the structured result, so it is shown as it arrived.

Your recent previews

What this does, and what it does not

The calendar is built in your browser from what you paste. Dates are normalised and a year-less date lands in the coming twelve months. Each row is typed into the four buckets of a catalyst calendar - Earnings (results, investor days, pre-announcements, debt maturities), Corporate (regulatory decisions, trial readouts, M&A milestones, lock-ups, product events), Industry (conferences, trade shows, industry data) and Macro (FOMC, CPI, payrolls, central banks) - and given an impact: high for a binary event on a held name, for earnings on a large or unsized position and for the major macro prints; one step up for anything on a position of 5% or more; one step down for non-earnings events on names you do not hold. An impact you type yourself is never changed.

It knows nothing you did not paste: no earnings-date feed, no consensus, no prices. Earnings dates shift, so confirm them with investor relations. The preview is research preparation, not investment advice. Derived from the agent skill @anthropics/catalyst-calendar (anthropics/financial-services-plugins, Apache-2.0). The example companies are fictional.

What is coming up, and what does it mean for the book?

Calendar (free, in your browser, no AI)

No exact date

What the browser found

Since your last preview (free - dates moved, events added or gone)

Draft preview (free - date order, impact from the event type and your book)

This week

Next week

This week's key events

Next week

Position implications

Dates to verify

Gaps in the list

Impact changes

Browser flags, answered

Reconciliation with your calendar

Weekly preview note

Raw model reply

Your recent previews

If that did not work

What this does, and what it does not