Claim Audit
Connect a claim to its decision, evidence source, tested context, exclusions, counterevidence and residual uncertainty, then produce a defensible claim draft.
DO LESS. PROVE MORE.
Make fewer claims.
Prove the ones that matter.
A practical field guide for making consequential technical claims with clearer boundaries, stronger evidence and less false certainty.
THE BOOK
Confidence should come from evidence, not volume.
Complex technical work often loses credibility because teams claim more than their evidence can support. Do Less. Prove More. offers a disciplined way to define the claim, identify the proof obligation, test what matters and state what remains unknown.
It is written for people making or governing consequential technical decisions without pretending uncertainty has disappeared.
- Senior engineers, architects and AI/ML practitioners
- Technical product and program leaders
- Technology executives responsible for evidence-backed decisions
FIELD TOOLS
Start with one real decision.
The tools run locally in your browser and require no account. Start with the Claim Audit. Open the specialist tools only when the work calls for them.
Show 7 specialist tools
Test Sufficiency Check
Assess whether evidence covers the decision-relevant scenarios, failure modes, environment, reproducibility and untested scope.
AI Evaluation Scope Builder
Frame model version, intended task, evaluation population, metrics, thresholds, comparators, exclusions and what the evaluation actually establishes.
Proof Boundary Canvas
Separate proven conditions, unproven conditions, explicit exclusions and residual uncertainty around a decision.
Proof Discipline Check
Review claim precision, traceability, boundary discipline, reproducibility, uncertainty disclosure and decision fit by dimension.
Evidence Ladder
Classify available evidence by strength and determine what additional evidence would materially change the decision.
Residual Uncertainty Register
Capture what remains unknown, consequence if wrong, relevance to the present decision and the next evidence action.
Claim Chain Inspector
Break a conclusion into linked claims and test whether each inferential jump has an explicit basis.
SAVED WORK · LOCAL FIRST
Pick up where you left off.
Save tool states and reports on this device, reopen them later and continue without retyping. Nothing in this workspace is sent to a server.
Stored only in this browserUse Export before clearing browser data or moving to another device.
SELECTED WORK ITEM
Your saved analysis will appear here.
Reopening restores the inputs so you can continue the work.
EXPLORE MORE
Go deeper when you need to.
Essays & notes
Examples, counterexamples and evolving ideas around claims, proof obligations and residual uncertainty.
COMING SOONReader resources
Glossary, worksheets, checklists and working templates that extend the book.
COMING SOONMobile companion
A mobile companion for capture, review and deeper proof workflows. Coming soon.
EXPLORE PUPFISH SOFT →Speaking & teams
Executive briefings, workshops and facilitated claim-audit sessions for technical organizations.
GET IN TOUCH →THE COMPANION BOOK
Leading outcomes is the other half of the problem.
The Outcome Leader focuses on turning strategy into measurable outcomes through clarity, decision rights, mobilization and evidence.