Guides / Cookbooks

Cookbooks · Screening and digests · Beginner · about 10 min

Send a pre-open digest

Before the open every weekday, ask the same questions about a short list of index ETFs and post one message with all of them to Slack.

1. When to use it

   You want one morning read in the channel the desk already watches, not one message per symbol. A workflow asks the questions for every symbol on its list at a set New York time and sends a single digest.

2. The workflow

   Here is the definition. 08:30 runs before the open on the previous session's numbers; 16:45 would use the session that just closed.

   {
     "name": "Pre-open read with trend",
     "symbols": "SPY QQQ IWM DIA",
     "days": [
       1,
       2,
       3,
       4,
       5
     ],
     "at_time": "08:30",
     "model": "ora-fin1",
     "questions": {
       "up_week": {
         "type": "yes_no",
         "instructions": "Will the stock be higher in a week?"
       },
       "move2": {
         "type": "yes_no",
         "instructions": "Will the stock move more than 2% in either direction in 5 trading days?"
       },
       "above_50d": {
         "type": "yes_no",
         "ask": "C > sma(C,50)"
       }
     },
     "actions": [],
     "digest": {
       "do": "slack",
       "connection": "trading-desk",
       "when": "always"
     }
   }

   Each symbol is an ordinary request, the same one you can send yourself:

   {
     "model": "ora-fin1",
     "state": {
       "symbol": "SPY"
     },
     "questions": {
       "up_week": {
         "type": "yes_no",
         "instructions": "Will the stock be higher in a week?"
       },
       "move2": {
         "type": "yes_no",
         "instructions": "Will the stock move more than 2% in either direction in 5 trading days?"
       },
       "above_50d": {
         "type": "yes_no",
         "ask": "C > sma(C,50)"
       }
     }
   }

   Open this in the Playground

3. The real response

   The four per-symbol answers for the session of 2026-10-30:

   Symbol	up_week	move2	above_50d (rule)
   SPY	0.598	0.394	0.1128
   QQQ	0.561	0.36	0.0653
   IWM	0.526	0.322	0.2176
   DIA	0.567	0.367	0.2849

   And the digest those four answers make, in the layout a workflow posts (assembled here from the responses above):

   Pre-open read with trend · data as of 2026-10-30 · 0 of 4 symbols fired:
   • SPY: up_week 60%, move2 39%, above_50d 11%
   • QQQ: up_week 56%, move2 36%, above_50d 7%
   • IWM: up_week 53%, move2 32%, above_50d 22%
   • DIA: up_week 57%, move2 37%, above_50d 28%

   The read is mixed: all four lean slightly up on the week, while the rule C > sma(C,50) says each one closed under its 50-day average (beliefs from 0.0653 to 0.2849). One message shows the desk both.

4. The code

   You don't need code for a workflow. To build the same read yourself (for example inside your own scheduler), loop over the symbols and format the line:

   cURLPythonJavaScriptPHPGoJavaC#RubyCopy

   Requires curl 7.76+. Save as run.sh, then: bash run.sh

   curl -sS https://oramodel.com/v1/systemone \
     -H "Authorization: Bearer $JEV_API_KEY" \
     -H "Content-Type: application/json" \
     -d '{
     "model": "ora-fin1",
     "state": {
       "symbol": "SPY"
     },
     "questions": {
       "up_week": {
         "type": "yes_no",
         "instructions": "Will the stock be higher in a week?"
       },
       "move2": {
         "type": "yes_no",
         "instructions": "Will the stock move more than 2% in either direction in 5 trading days?"
       },
       "above_50d": {
         "type": "yes_no",
         "ask": "C > sma(C,50)"
       }
     }
   }'

   import os
   import requests
   
   URL = "https://oramodel.com/v1/systemone"
   HEADERS = {"Authorization": f"Bearer {os.environ['JEV_API_KEY']}"}
   
   SYMBOLS = ["SPY", "QQQ", "IWM", "DIA"]
   QUESTIONS = {
       "up_week": {
           "type": "yes_no",
           "instructions": "Will the stock be higher in a week?"
       },
       "move2": {
           "type": "yes_no",
           "instructions": "Will the stock move more than 2% in either direction in 5 trading days?"
       },
       "above_50d": {
           "type": "yes_no",
           "ask": "C > sma(C,50)"
       }
   }
   
   lines = []
   for s in SYMBOLS:
       r = requests.post(URL, headers=HEADERS, timeout=30,
                         json={"model": "ora-fin1", "state": {"symbol": s}, "questions": QUESTIONS})
       r.raise_for_status()
       a = r.json()["answers"]
       lines.append(f"• {s}: " + ", ".join(f"{k} {v['yes_no']:.0%}" for k, v in a.items()))
   print("\n".join(lines))

5. Act on it

   1. Go to Workflows → New workflow, or start from the Pre-open market read template under Examples → Workflows.
   2. Set the symbols, weekdays and 08:30, paste the questions, and set the digest to Slack → trading-desk, always.
   3. Save (it checks everything with a dry run), then Test run: you see each symbol's answers and the digest, and nothing is sent.
   4. Untick dry run. Holidays and late data are skipped, so the digest never repeats an old session.

6. Variations

   • Add a per-symbol action (for example up_week >= 0.62) and set the digest to fired, so the message lists only the names that crossed the line.
   • Send the digest to a webhook instead: the body is {event: "jev.workflow", workflow, run_id, as_of, symbols, answered, fired_symbols, rows} (Workflows reference).
   • Run after the close at 16:45 on the session that just ended.

Information, not investment advice. Every number on this page is a real response for the session shown; answers change with each session.

What's next

• Run a watchlist every day after the close
• Rank a watchlist by probability