The AI prompt I run before every sell decision

I’ve held META through three successive raises to its capex guidance, capex being the money a company spends building things, in META’s case the data centres behind its AI push. The third raise, in Q1 2026, pushed the band to $125–145bn for the full year, well clear of the $120bn peak I’d assumed when I bought. For the first time, I ran an AI prompt before the sell decision, a structured thesis audit rather than a gut check. The model’s read: the original bounded-capex thesis is gone. What I’m doing about that (hold, sell some, or exit) is my decision, and a separate question from whether the prompt did its job.

That separation matters. The point of the prompt is not to maximise returns on any single holding. It’s to make sure I’m not making sell decisions on price when the real question is whether my reason to own still holds. The output is a defensible thesis read, not a trade.

Every sell-decision prompt I’ve seen on the internet asks AI to recommend whether to sell. That’s the wrong question. The right question is whether the reason you bought still holds. Those are different questions, and only one of them is something AI can answer honestly.

What most sell-prompt content gets wrong

When I search “AI prompt for selling stock” I get one of two things. Either generic exit-strategy generators (“provide an exit strategy for [company], define selling signals such as valuation peaks or earnings disappointments”) that invent the signals for you. Or roundup posts where one of seven prompts is a vague “act as a financial adviser and tell me whether to hold.”

The standard mistake is asking AI for a recommendation. AI hedges recommendations: “it depends on your risk tolerance, your time horizon, and current market conditions.” That’s the response the model is built to give. True, unfalsifiable, and no help whatsoever to someone with a real position and a finger over the sell button.

The harder part of the sell decision isn’t getting a recommendation. It’s checking whether the thesis you built when you bought still holds. That’s a reasoning task: the input is your original thesis, the output is whether it survives what has happened since. AI is good at that, once you ask the right question.

The standard sell-prompt asks AI to recommend. This one asks AI to audit.

The prompt has four sections (SCOPE, FILTER, RISK, VERDICT) following the Prompt Stack structure. The SCOPE keeps the model auditing your thesis rather than handing you a recommendation. The FILTER forces you to type your original thesis before reading any output. The RISK names the rationalisation you are most likely to make. The VERDICT closes with thesis-anchored action, not price-anchored action.

The AI thesis-audit prompt

I run this prompt every time I notice a position straining. The FILTER section forces me to commit the original thesis in writing before reading any output, and that act is most of the work. It follows the Prompt Stack: SCOPE, FILTER, RISK, VERDICT. The placeholders in square brackets are yours to fill in. The FILTER section is where the work happens: if you can’t state your original thesis in three sentences, the problem is the thesis, not the prompt.

The SCOPE keeps the model auditing the thesis rather than handing you the recommendation it wants to give. The FILTER forces you to type your original thesis before reading anything the model returns. That act is the behavioural intervention. The VERDICT closes with thesis-anchored action, not price-anchored action. Price-anchored: “it’s down 12%, cut your loss.” Thesis-anchored: “the free-cash-flow thesis still holds, but capex guidance is a legitimate flag: sell half.” (Free cash flow: the cash a business has left after running itself and building for the future. It’s what funds the dividend, the buyback, or the next bet.)

What it caught on META

Here is the FILTER section I ran on META the morning of 2026-05-01, after the post-earnings drop.

• Position: Long META, held since late 2024 / early 2025, average price paid roughly $500, current price around $610.
• Original thesis: META is investing through an AI infrastructure cycle. Ad revenue growth is durable. The free-cash-flow trajectory makes the valuation defensible if capex is bounded. My assumption when I bought was annual capex would peak somewhere around $120bn.
• What has happened since: FY25 came in at $72.2bn, fine, but management opened FY26 guidance at $115–135bn in January, then raised it again at Q1 results in April to $125–145bn. Three guidance updates, each one higher. The top end of the latest band is $145bn, over 20% above the $120bn peak I’d assumed when I bought.

Question 1 made the standard case for AI investment: ad revenue growth is intact, the spend gets absorbed as revenue grows, and the squeeze on free cash flow is a timing problem, not a permanent one.

Question 2 was the one that did the work. The model wrote, in effect: “Three successive moves up break the bounded-capex assumption, not just strain it. The original thesis treated $120bn as a peak. A peak that gets raised the very next quarter isn’t a peak, it’s a floor. The FCF trajectory you bought is not the FCF trajectory you own.” That was sharper than the sentence I would have written for myself. It made the inference rather than restating the input, the difference between “capex is rising” and “the assumption your thesis rested on is gone.”

The RISK call landed harder still. The rationalisation it named: “You will tell yourself the AI investment thesis justifies the capex, which is exactly what management is also telling you.” That’s the sentence you read and wince at, because you’ve already started constructing the justification before you finished the FILTER.

What I do with this read is a separate question from whether the prompt did its job. The bounded-capex assumption is gone. That’s the prompt’s conclusion in writing, the discipline working. The decision the prompt won’t make for me: hold because the AI capex spend may yet justify itself, sell the thesis-broken portion, or exit because the thesis I bought no longer applies. Mine to weigh.

The prompt's job ends at the thesis read. The trade is yours.

Three weeks on, I ran a simpler version through Claude (Opus 4.7, with live web search, 22 May 2026), sell-some-vs-hold on the same META holding, with the same capex-raise context. The prompt’s discipline still holds. Claude reframed the bounded-capex break in sharper language than my original Q2 paraphrase: “the floor of 2026 guidance now sits above the ceiling you assumed.” The case for selling some, in Claude’s read: the part of the spending driven by rising memory-chip prices is “the worse kind of capex increase, because you’re paying more for the same capacity rather than buying more capacity. That’s a margin signal, not a growth signal.” The honest caveat in the same response: “selling a compounder into capex fear is historically how people have given up Meta’s best runs.” Different framing of the question; the same fork.

Where the prompt falls short

Six caveats worth being honest about up front. I learned most of these the hard way running this on real positions; worth reading before you run it on yours.

It requires the original thesis in writing. If you cannot state in three sentences why you bought, the prompt returns a generic audit of a stock you’ve described, not an audit of your thesis. Most retail holders don’t have their entry thesis written down. The real value is forcing them to write it.

The honest read depends on what you put in the FILTER. If you summarise the negative evidence accurately, the model produces a useful synthesis. If you frame it charitably, the model will too.

AI cannot audit what you haven't disclosed.

It doesn’t tell you how much to sell. The VERDICT gives you “hold, sell some, or exit”, not “sell exactly 30%.” How much depends on how big the holding is, the price you paid, and the rest of your portfolio. The prompt is a thesis check, not a sizing tool.

It is most useful before price pressure, not under it. Running it after the stock is down 15% is better than not running it, but the RISK stage is harder to take seriously when you’re already down. The discipline is to run it at a calm moment, when you notice the thesis is straining.

AI treats your stated thesis as correct. If your original thesis was wrong from the start, the model audits it on its own terms without flagging the original error. It audits consistency, not correctness. The 5 questions to ask AI before buying any stock covers the other side of the discipline.

It’s a thesis check, not a portfolio tool. Some sell decisions aren’t thesis-driven at all: selling some because a position got too big, raising cash, or cases where the thesis is fine but the money would do more work elsewhere. The prompt doesn’t help with any of those. Confusing the two produces worse decisions than running no prompt at all.

The sell decision is not the hard part. Recalling what you were actually betting on, and being honest about whether that has changed, is the hard part. The prompt is the discipline that makes you do the recalling before the price tells you what to think. It won’t place the trade for you, which is just as well. That part should still cost you a little sleep. The Prompt Stack covers the underlying methodology, free and ungated. The buy side of the same discipline is at 5 questions to ask AI before buying any stock. Every documented failure from running these prompts on real positions is collected on lessons. The three real exits this prompt ran against (CRM, INTU, TSLA) are at /trades.