Bayesian thinking · IN SHORT
Record what you believed before a new signal, then examine how much that signal should change your view. The Bayesian Thinking skill helps make the update visible without inventing numerical certainty.
Your brief
Three fictional prospects request an export feature, but all came through one partner. That supports interest within this group without establishing demand throughout the market.
Possible draft
Confidence may rise slightly. A useful next test is a small prototype shown to prospects outside the partner channel, with a decision rule agreed before the test.
01 / THE FOUNDATION
The output records the prior belief, supporting and opposing evidence, a proportionate update and a next action that can still be revised.
02 / SEE HOW IT WORKS
Three fictional prospects request an export feature, but all came through one partner. That supports interest within this group without establishing demand throughout the market.
Confidence may rise slightly. A useful next test is a small prototype shown to prospects outside the partner channel, with a decision rule agreed before the test.
The skill organises evidence. It is not a validated statistical model and cannot supply missing probabilities to make an answer look precise.
03 / GET STARTED
State the claim and what would count against it before considering the new evidence.
Check reliability, independence, sample size and alternative explanations.
Describe how confidence moves and choose the next observation that could change the decision.
04 / MAKE IT YOURS
Download the ZIP for free. The English ZIP contains the SKILL.md instructions and is available to download without an account. This archive does not state a licence; check the permitted reuse terms before adapting or redistributing it. Need help applying this skill? Book a call.
Full informationDownload. Apply. Refine.
File language: English
Explore the details at your own pace.
One encouraging signal should not make a weak plan certain, and one bad week should not erase years of evidence. This free Claude skill makes your starting belief visible, weighs what the new evidence can actually tell you and records how far your confidence should move.
A fixed route keeps the work reviewable. Each step produces input for the next one.
People update beliefs all the time, but usually hide the starting point. A new lead feels like proof that a market exists; a lost deal feels like proof that it does not. Without a stated prior, both reactions can be defended after the event. The skill asks what you believed before the signal and what evidence created that belief.
A prior does not need a decimal. High, medium or low confidence with a reason is often more honest than 63 percent. When credible base-rate data exists, Claude can use it. When it does not, the skill keeps the estimate qualitative and names what would make it stronger.
Five articles repeating one press release are not five independent signals. A customer interview is relevant to that customer but may not describe the market. A conversion result from a tiny sample can be promising without being stable. The skill separates source reliability, directness, independence and sample quality before it updates anything.
It also asks what else could explain the observation. A campaign may improve because demand rose, tracking changed or the offer became cheaper. Naming alternatives does not cancel the result. It stops one convenient explanation from becoming the only explanation.
you > /ai-bayesian-thinking-coach claude > intake → method → checked output [ok] assumptions remain visible
The answer shows the original claim, prior, strongest supporting evidence, strongest opposing evidence and the resulting confidence. It explains the movement in words. If you supplied real probabilities and conditional evidence, it can show the calculation, but it never manufactures a likelihood just to make the answer look scientific.
The final line is practical: keep the decision, change it, pause it or run a test. Claude also proposes the next observation with the highest information value. The point is not to sound uncertain forever. It is to spend uncertainty on the right experiment.
The skill works well for product demand, marketing signals, hiring assumptions, project risks and forecasts that will receive more evidence. Give it the claim, the starting view and the new information. If different people hold different priors, record them separately before discussing the update; that disagreement is useful evidence about the decision process.
It is not a substitute for a regulated professional, a statistical model or a risk system. Medical, legal and financial decisions need qualified review and validated data. The skill can organise questions for that review, not replace it.
Suppose three prospective customers ask for the same feature. The prior is that the feature may help a narrow segment but is not yet a market requirement. The signal supports demand, but all three leads came through one partner and may not be independent. Confidence moves upward a little, not to certainty.
The useful next step is not a full build. It is a short prototype shown to customers outside that partner channel, with a predefined decision rule. The update has changed the action while keeping the uncertainty visible. That is Bayesian thinking in ordinary work.
It can calculate an update when you provide defensible probabilities. Otherwise it uses a qualitative update and explains why a precise number would be misleading.
Claude helps you recover one from base rates, previous decisions and what you would have predicted before seeing the result. It labels the reconstruction as an estimate.
Yes. Ask every decision-maker to state a prior independently, then compare what evidence would change each view.
No. The action also depends on cost, reversibility, timing and risk. The skill keeps those separate from the probability of the claim.
Bring one concrete process, page or decision. We will tell you what can be improved, what should stay human and what evidence is still missing.
