Claude Fable 5 for the hardest long-running work

Plan a staged migration from the legacy billing service to the new event-driven architecture using the repository, incident history, architecture notes, and test suite.

First map dependencies, invariants, data ownership, failure modes, and unknowns. Then propose reversible milestones with acceptance tests and rollback criteria. Do not change production systems or approve irreversible actions.

At each checkpoint, update a decision log and risk register. Finish with the smallest reviewable implementation package and separate summaries for engineers and executives.

Evaluate whether our B2B product should enter Germany, France, or the Netherlands using the supplied market studies, interviews, competitor materials, regulatory notes, and operating assumptions.

Build an evidence map before reaching a conclusion. Separate verified facts, source claims, inference, and open questions. Identify contradictions, test alternative explanations, and show what evidence could reverse the recommendation.

Deliver a market comparison, three scenarios, a recommendation with confidence, and a 30-day validation plan. Cite every material claim to the supplied source.

Audit the attached operating model and board memo. Trace the revenue, cost, cash, and headcount assumptions through every material formula and reconcile the workbook with the narrative.

Flag broken formulas, hidden constants, inconsistent periods, unsupported assumptions, circular logic, and places where charts conceal uncertainty. Run sensitivity cases for the five assumptions with the largest decision impact.

Do not provide investment advice. Return an issue register, corrected logic proposals, scenario table, and questions the finance owner must answer before approval.

Prepare a board decision package for the proposed acquisition using the diligence room, management interviews, customer analysis, financial model, legal red flags, and integration plan.

Reconcile disagreements across sources. Separate facts, management claims, assumptions, and unresolved risks. Compare acquire, partner, and do-nothing scenarios using the approved decision criteria.

Produce an executive memo, decision matrix, risk register, integration dependencies, questions for management, and an appendix that traces each conclusion to evidence.

Coordinate the launch of our new creator product across the US, UK, Germany, and Japan using the approved product claims, research, brand system, channel plan, localization rules, budget, and legal restrictions.

Create one source-of-truth launch plan with dependencies, owners, review gates, market differences, message architecture, and measurable success criteria. Flag claims or cultural assumptions that need local validation.

Then prepare connected briefs for landing pages, paid ads, UGC concepts, social assets, product visuals, launch videos, voiceover, and music production in Neurohelper’s specialist modules.