COMMENT "OVERNIGHT"
The overnight market-scan prompt (ranked watchlist by morning)
One prompt that runs while you sleep: it scans the last 24h of movers and catalysts, scores each setup, and hands you a ranked top 5 with entry, stop and target. Save it as a Grok Task to run nightly, or paste it into Claude/ChatGPT before bed.
01Quick steps
- 1In Grok, save the prompt (below) as a daily Task so it runs on its own; in Claude/ChatGPT, paste it before bed.
- 2It scans top movers and catalysts in US stocks and crypto from the last 24h.
- 3It scores each setup on momentum, volume and news and returns a ranked top 5.
- 4Each comes with an entry, a stop and a target, plus one line on why.
- 5In the morning, trade only the ones you actually understand. Crypto names go on Photon (https://photon-sol.tinyastro.io/@ttplus10builds) with the scan's stop and target typed in; stock names go through your broker.
02Your ready to use prompt
Every night, scan the top movers and biggest catalysts in US stocks and crypto from the last 24 hours. Score each setup 1 to 10 on momentum, volume and news. Return a ranked top 5 watchlist with an entry, a stop loss and a target for each, plus one line on why. Skip anything with no real catalyst.
03Full tutorial
The bottleneck in trading is reviewing the whole market every day. This prompt offloads the scan: overnight it reviews 24h of movers and catalysts, scores them, and hands you a short ranked list with levels, so you wake up to a plan instead of a blank chart.
In Grok, save it as a daily Task (Grok runs recurring tasks on its own and can pull live data). In Claude or ChatGPT, paste it manually before bed. The Grok Task route is the 'while you sleep' version because it fires without you.
Use the script above. Tune the universe (US stocks, large-cap crypto, or a specific sector), the scoring weights (momentum vs volume vs news), and the 'skip anything with no real catalyst' filter so it does not hand you low-float noise. The sharper the constraints, the better the list.
Force it to return an entry, a stop and a target for each name, plus the one-line catalyst. A watchlist with no levels is useless; levels make it actionable and let you size risk per trade.
In the morning, cut the list to the few setups you actually understand and that have a real catalyst. The AI ranks; you decide. Trading a name you do not understand because a model scored it 8/10 is how you give the edge back.
For each trade you take, risk a fixed small percent and use the stop the scan gave you. For the crypto names, place them on Photon (https://photon-sol.tinyastro.io/@ttplus10builds) and enter that stop and target before you confirm. The prompt finds setups; your risk rules keep one bad morning from mattering.
A scanner is an idea generator, not a crystal ball. Most people lose money trading; only risk what you can afford to lose. Not financial advice.
A scanner surfaces ideas, it does not guarantee them. Only trade setups you understand and risk what you can afford to lose. Not financial advice.