01
AI quality of earnings review: 4 prompts to find the real profit
Headline profit isn't always what a company earned.
Prompt Stack 10 min read
Posts · 7 in this category
Prompt Stack
One paragraph that stops AI collapsing facts and guesses into the same confident answer.
The evidence so far
1 AI errors and 4 catches logged in 3 of these 7 posts, up to 16 Jun 2026.
02
AI earnings call red flags: three phrases to watch for in the transcript
AI earnings call red flags: three patterns that recur across calls.
Prompt Stack 9 min read
03
Claude prompts for investing: 6 real examples
Six Claude prompts tested on MSFT, META and NVDA, with their real outputs and the source audit that found the NVDA input itself was wrongly labelled.
Prompt Stack 16 min read
04
The one AI prompt I run the morning before earnings
Most AI earnings prompts are reactive.
Prompt Stack 11 min read
05
AI earnings call analysis: the 5 prompts I actually run
Five prompts for AI earnings call analysis that read the language the numbers miss: omissions, performative confidence, and quarter-on-quarter drift.
Prompt Stack 14 min read
06
The single prompt change that made AI analysis worth using
One AI prompt separates observable facts from assumptions in any stock analysis.
Prompt Stack 10 min read
07
7 AI prompts for covered calls (2026)
Seven AI prompts for covered calls that start from verified broker data and use the model as a decision reviewer, not a source of market prices.
Prompt Stack 24 min read
No posts match that.
Try another word, or show every category.
About Prompt Stack
Posts here apply the Prompt Stack, the four-stage structure I run before I trust AI with a decision that matters: SCOPE, FILTER, RISK, VERDICT. The full method, and the reasoning behind each stage, lives at /prompt-stack. These posts are the method at work on real decisions, most of them about money, where a wrong answer costs something.
The pattern in each one is the same, whatever the decision. A real one I was facing (an earnings release I held through, a covered call I was about to sell), then the exact prompt I ran, what the model returned, and where the output still needed checking. The prompts are written to be copied; square-bracket placeholders mark what you fill in.
If you're new to the method, the single prompt change is the right entry point: one paragraph that stops AI collapsing facts and inferences into the same confident answer. The rest of these posts build on that idea, one decision at a time.
Free letter
Confidently Wrong, every fortnight.
Which AI told the truth this month, with the receipts, plus an AI story worth retelling in between.
Free. Leave whenever you like.