Claude Fable 5.1 Prompt Generator

I have a Node.js service that intermittently returns HTTP 500 errors under load. The error occurs in approximately 0.4% of requests. I've included the full service code, the last 500 error log lines from a production incident, and the database schema. I do NOT want a surface fix. I want: 1. ROOT CAUSE — trace the exact failure path from the triggering condition to the 500 response 2. WHY INTERMITTENT — explain what race condition, resource contention, or timing dependency makes this fail only sometimes 3. THE FIX — the minimal, correct change with the exact file and line number 4. VERIFICATION — a specific test case that reproduces the original bug and passes with the fix 5. MONITORING — the alert rule that would have surfaced this within the first 10 occurrences [Paste service code and error logs]

Design a production-grade multi-agent workflow for the following task: process 500 legal contracts per day — extract key terms, flag non-standard clauses, summarize obligations per party, and generate a risk-ranked review queue for legal staff. Deliver: 1. AGENT ARCHITECTURE — DAG of all agents; which run in parallel vs. sequential; how output from one feeds the next 2. PROMPT TEMPLATES — exact system prompt and user prompt for each agent (minimum 5 agents) 3. OUTPUT SCHEMAS — typed JSON schema for each agent's output 4. ORCHESTRATION LOGIC — how the coordinator decides when to re-run a sub-agent vs. accept imperfect output 5. COST MODEL — estimated per-contract token cost at Claude Fable 5.1 pricing ($7.50/MTok input, $37.50/MTok output) vs. Fable 5 ($10/$50/MTok), with cache-read savings quantified at the 75% discount rate 6. FAILURE HANDLING — what happens when a contract is malformed, too long, or uses a foreign language This will run as a SaaS product used by law firms.

Perform an Enterprise Frontier-grade security architecture review of the following system: a B2B SaaS platform handling PII for 2M users, processing financial transactions, and integrating with 12 third-party APIs. I'm sharing the full system architecture diagram, all API integration specs, the data flow documentation, and the current security controls list. Deliver: 1. THREAT MODEL — STRIDE analysis across all components; enumerate every threat actor, attack surface, and data flow risk 2. CRITICAL VULNERABILITIES — the 5 highest-severity findings with CVSS score, attack path, and proof-of-concept scenario 3. COMPLIANCE GAPS — gaps against SOC 2 Type II, PCI DSS 4.0, and GDPR Article 32 4. PRIORITIZED REMEDIATION — P0/P1/P2 remediation plan with engineering effort estimate and implementation sequence 5. MONITORING ARCHITECTURE — the specific log events, anomaly thresholds, and alert rules that would detect each critical threat [Paste architecture docs]

I'm sharing the full codebase of a 6-year-old fintech application (approximately 250,000 lines of code, Python/Django). The engineering team says 'it's held together with duct tape.' I need an honest, thorough technical debt assessment — not a list of cleanup tasks, but a prioritized plan that connects technical debt to business risk. Deliver: 1. DEBT INVENTORY — categorize all technical debt found in the codebase by type (design, code, test, documentation, infrastructure) and severity 2. BUSINESS RISK MAPPING — for each P0/P1 debt item, state the specific business failure mode it creates (e.g., 'this race condition in the payment reconciliation service will produce silent over-billing at >500 concurrent transactions') 3. DEPENDENCY GRAPH — which debt items block others; what must be resolved first 4. 6-MONTH PAYDOWN ROADMAP — sprint-by-sprint plan that pays down debt while keeping the product shipping 5. NON-NEGOTIABLES — debt items that are safety/compliance risks and cannot be deferred under any circumstances [Paste codebase]

I'm attaching 12 peer-reviewed papers on GLP-1 receptor agonist use for obesity management in adolescents (published 2021–2026). Synthesize them as a clinical practice guideline draft: 1. EVIDENCE SUMMARY — what the 12 papers collectively establish with high confidence; what remains contested 2. CONTRADICTION MATRIX — every direct factual contradiction between papers; explain the methodological differences that account for each 3. DOSING EVIDENCE — what the evidence says about dosing regimens, titration schedules, and duration; note where evidence is weak 4. SAFETY PROFILE — adverse events reported across all 12 papers; calculate pooled rates where data is sufficient 5. POPULATION GAPS — which patient subgroups are underrepresented or absent from the evidence base 6. CLINICAL RECOMMENDATIONS — draft 6 evidence-graded recommendations (A/B/C/D per GRADE) based only on what the papers actually establish Do not extrapolate beyond the evidence. Flag every claim with its source paper. [Paste paper texts]

Design the complete technical architecture for a greenfield B2B analytics SaaS product. Requirements: 10,000 business customers, each with up to 1M data rows per day; real-time dashboards with <200ms p95 query latency; multi-tenant data isolation; SOC 2 Type II target within 18 months; starting with a team of 8 engineers. Deliver: 1. SYSTEM DESIGN — full architecture diagram in text/ASCII; all components, data flows, and integration points 2. TECHNOLOGY CHOICES — for each major component (API layer, data warehouse, streaming, caching, auth, CDN), give the chosen technology, 2 alternatives, and why you chose this one 3. MULTI-TENANCY MODEL — exactly how data isolation is implemented at the database and API layer; the tradeoffs 4. SCALABILITY PLAN — how the architecture handles 10× growth without a rewrite 5. SOC 2 READINESS MAP — which controls are built-in vs. need to be added; the 6-month path to audit readiness 6. TEAM STRUCTURE — squad topology for 8 engineers building this architecture 7. WEEK 1 PLAN — the exact sequence of technical decisions and first commits the team makes on day 1

I'm an investor conducting technical due diligence on a Series B ML infrastructure startup ($18M ARR, $95M valuation ask). I'm sharing their full GitHub organization (all repos), technical documentation, architecture overview, and customer interview transcripts. Conduct a senior engineering-level technical due diligence and deliver: 1. TECHNOLOGY ASSESSMENT — honestly rate the codebase quality, technical debt level, and architecture soundness (1–10 with evidence) 2. MOAT ANALYSIS — is their technical advantage real, durable, and defensible, or is this replicable in 6 months by a well-funded competitor? 3. SCALABILITY CEILING — at what revenue/customer volume does the current architecture break, and what's the cost to fix it? 4. TALENT RISK — from the commit history and code quality patterns, assess the engineering team's strength and key-person risk 5. RED FLAGS — the 3 most important technical concerns that should affect valuation or deal structure 6. INVESTMENT THESIS IMPACT — given the technical findings, what adjustments to the investment thesis or deal terms do you recommend? [Paste repo access and docs]

I'm advising a healthcare technology company on the regulatory implications of deploying a clinical decision support AI system in the US, EU, and Canada. The system analyzes patient data and recommends treatment protocols. Deliver a complete regulatory analysis: 1. US REGULATORY PATH — FDA classification (SaMD risk class), predicate device strategy vs. De Novo, required clinical evidence, timeline, and estimated cost 2. EU PATH — EU AI Act risk classification, CE marking requirements under MDR 2017/745, notified body process, and Article 22 data governance requirements 3. CANADA PATH — Health Canada SaMD pathway, clinical trials requirements, and provincial health authority considerations 4. DATA PROTECTION — HIPAA (US), GDPR (EU), PIPEDA (Canada) requirements specific to AI-generated clinical recommendations and the 'right to explanation' 5. CONFLICT MATRIX — where the three regulatory regimes directly conflict and how to navigate the conflicts 6. DEPLOYMENT SEQUENCE — in which market to seek approval first, and why Assume the system processes de-identified data with re-identification risk.

I'm sharing my entire Python monolith (300,000 lines, SQLAlchemy 1.4, Flask 2.1, Celery 5.2) that needs to migrate to an async FastAPI + SQLAlchemy 2.0 architecture. The migration must be zero-downtime. Deliver a complete migration plan: 1. DEPENDENCY GRAPH — map all current components and their inter-dependencies; identify the safe extraction sequence 2. MIGRATION SEQUENCE — which modules to migrate first, with explicit rationale based on coupling and risk 3. DUAL-WRITE STRATEGY — exactly how to run old and new systems in parallel during transition without data loss or inconsistency 4. ASYNC CONVERSION GUIDE — the specific patterns to convert SQLAlchemy 1.4 sync sessions to 2.0 async sessions, with before/after code examples for the 5 most common patterns in this codebase 5. ROLLBACK PLAN — for each migration phase, the exact steps to roll back if the migration fails in production 6. TESTING STRATEGY — the test coverage required before cutting over each module 7. TIMELINE — realistic 6-month milestone plan with go/no-go checkpoints [Paste codebase]

I'm the CEO of a US-based B2B SaaS company ($8M ARR, 120% NRR, HR tech sector) considering expansion into the German market. Deliver a complete market entry strategy: 1. MARKET SIZING — German HR tech TAM/SAM/SOM with methodology; how it compares to the US market in buying behavior and willingness to pay 2. REGULATORY LANDSCAPE — Works Council requirements (Betriebsrat), Bundesdatenschutzgesetz (BDSG), GDPR specifics for HR data in Germany, and what product changes are legally required before selling to German companies 3. COMPETITIVE MAP — the top 6 German/EU HR tech competitors, their positioning, pricing, and the gaps our product could exploit 4. GTM MOTION — direct sales vs. reseller vs. partnership with a German HR consultancy; optimal sequence and economics for each 5. HIRING PLAN — the exact first 3 hires in Germany (title, role, why in that order) and the legal entity structure 6. FINANCIAL MODEL — break-even analysis at 3 GTM scenarios; how long until the German operation is self-funding 7. RISK REGISTER — the 5 highest-probability failure modes for US SaaS companies entering Germany, and mitigation for each

Write chapters 1–3 of a literary novel set in contemporary Lagos and London. The dual-POV structure alternates between: - YEMI: 41, a Nigerian-born cardiologist at King's College Hospital, London. Her chapter opens on the morning she receives a WhatsApp message that her estranged father has died and left her his debt-ridden private clinic in Lagos. - CHUKWUEMEKA: Yemi's 19-year-old nephew in Lagos, running the clinic's informal operations since the father's final illness. Requirements: - 10,000–12,000 words total across three chapters - Chapters alternate POV (Yemi Ch1 → Chukwuemeka Ch2 → Yemi Ch3) - Each chapter must establish its POV character's specific relationship with failure — but show it through action and scene, never state it - Dialogue must carry the register difference between diaspora English (Yemi) and Lagos street English (Chukwuemeka) without resorting to phonetic spelling - The tension between the two settings (NHS order vs. Lagos improvisation) must be structural, not ornamental - No backstory dumps. Each chapter drops the reader into a scene.

I'm about to negotiate a SaaS Master Service Agreement with a Fortune 500 customer. They've sent their standard vendor MSA. I'm attaching the full contract (82 pages). Prepare me for the negotiation: 1. POWER ANALYSIS — assess the contract's overall favorability balance (1–10 scale); which party wrote it with their interests in mind? 2. WALK-AWAY CLAUSES — identify every clause that, if left unchanged, would make this contract a material business risk for a Series B SaaS vendor. These are non-negotiable. 3. HIGH-LEVERAGE NEGOTIATION POINTS — the 8 clauses most likely to have negotiating room based on standard enterprise SaaS deals; suggest the ask for each 4. CONCESSION STRATEGY — which points can I concede without real business impact, ordered by the impression of value they create for the buyer 5. RED-LINE LANGUAGE — for the 5 most critical clauses, provide replacement language I can submit in a tracked-changes redline 6. NEGOTIATION SEQUENCING — the order in which to raise issues to maximize overall outcome [Paste contract]

I'm sharing the diff for a large pull request adding an AI-powered recommendation engine to our e-commerce platform. The PR adds approximately 4,200 lines of code across 23 files. Conduct a thorough senior-engineer code review: 1. CORRECTNESS BUGS — every bug where the code will produce wrong results in a real scenario. Include the exact input that triggers it and the wrong output it produces. 2. SECURITY VULNERABILITIES — every security issue, with severity (Critical/High/Medium/Low) and the attack vector 3. PERFORMANCE ISSUES — every N+1 query, missing index, unoptimized hot path, or memory leak, with the specific load scenario where it causes degradation 4. ARCHITECTURE CONCERNS — structural issues that will cause problems at scale or that create technical debt 5. MISSING TEST COVERAGE — specific scenarios that need tests before this ships 6. INLINE NITS — specific line-by-line suggestions (variable names, unnecessary complexity, missing error handling) For each finding: file name, line number, the problem, and the fix. [Paste PR diff]

Build a complete 5-year financial model for a B2B SaaS company with the following metrics: $4.2M ARR growing at 85% YoY, 115% NRR, $28K ACV, 9-month payback, 72% gross margin, -$180K monthly EBITDA, $3.1M cash runway. Deliver: 1. BASE MODEL — month-by-month P&L, cash flow, and headcount plan for 5 years; show all growth, churn, and cost assumptions explicitly 2. THREE SCENARIOS — bear (40% growth), base (85%), bull (130%); for each show the cash-out date, break-even month, and 5-year ARR 3. UNIT ECONOMICS EVOLUTION — how CAC, LTV, payback, and gross margin should evolve as the company scales; flag where current metrics signal concern 4. FUNDRAISING MILESTONES — the specific ARR and burn rate targets the company needs to hit to raise a Series A at a $40M+ pre-money valuation 5. HIRING PLAN — the optimal sequence of the next 20 hires across sales, engineering, CS, and G&A; justify each hire with the revenue it unlocks 6. SENSITIVITY TABLE — the 5 variables that most affect the 5-year outcome; show the model output range for ±20% on each Show all formulas and assumptions explicitly.

Conduct a comprehensive EU AI Act compliance audit for an AI system used in employee performance management (automated scoring of sales representatives based on call recordings, CRM activity, and manager ratings). Deliver: 1. RISK CLASSIFICATION — classify the system under EU AI Act Annex III; justify the classification with specific provisions 2. PROHIBITED PRACTICES CHECK — evaluate whether any components of the system qualify as prohibited AI practices under Article 5 3. HIGH-RISK REQUIREMENTS MAPPING — for each Article 9–15 requirement, assess the current system's compliance status (Compliant / Gap / Requires Assessment) and identify what evidence or controls are needed 4. FUNDAMENTAL RIGHTS IMPACT ASSESSMENT — assess the system's impact on workers' rights to dignity, non-discrimination, and data protection 5. CONFORMITY ASSESSMENT PATHWAY — the specific steps required to achieve conformity, in sequence 6. TIMELINE AND COST — realistic timeline and resource estimate to achieve compliance from current state Assume the system is already deployed and the company has 6 months before enforcement.

I'm the Head of Product at a Series B SaaS company. I have 340 feature requests in our backlog, 6 months until our next major customer review cycle, and a team of 12 engineers. I'm sharing our backlog export, customer interview transcripts (28 interviews), and our OKRs for the next 6 months. Build a strategic product roadmap: 1. THEME EXTRACTION — from the 28 interviews and backlog, identify the 5 true customer outcome themes (not feature requests); show how the feature requests map to each theme 2. PRIORITIZATION FRAMEWORK — apply ICE (Impact × Confidence × Effort) scoring to the top 40 requests; show your scoring methodology 3. 6-MONTH ROADMAP — a month-by-month delivery plan for 12 engineers; show which themes each sprint advances 4. TRADE-OFF MEMO — the 3 most painful prioritization decisions, the alternatives considered, and why this sequence beats them 5. RISK REGISTER — the 5 things most likely to blow up this roadmap; mitigation for each 6. SUCCESS METRICS — specific, measurable outcomes at 3 months and 6 months that prove the roadmap is working 7. STAKEHOLDER NARRATIVE — a 3-paragraph executive summary that explains the strategy to the board without using product jargon [Paste backlog and transcripts]

I'm an investigative journalist. I've gathered 4,000 pages of corporate documents, internal emails, regulatory filings, and deposition transcripts related to an alleged environmental cover-up by a chemical manufacturer. I'm sharing the full document set. Analyze the documents as an experienced investigative editor would: 1. DOCUMENT MAP — catalog all documents by type, date, and key parties; build a timeline of events 2. KEY FINDINGS — the 10 most significant facts these documents establish; ranked by newsworthiness 3. CONTRADICTION ANALYSIS — every place where internal documents contradict public statements, regulatory filings, or sworn testimony; cite the exact documents and pages 4. SOURCE RELIABILITY ASSESSMENT — for each document type, assess its authenticity indicators and what a defense lawyer might argue about admissibility or provenance 5. LEGAL EXPOSURE ANALYSIS — based on what the documents show, what laws may have been violated and by whom 6. STORY ARCHITECTURE — how to structure a 3-part investigative series that leads with the most compelling evidence while protecting sources 7. GAPS — what documents are missing that, if obtained, would clinch or collapse the story [Paste documents]

Design the integration architecture for connecting 8 enterprise systems: Salesforce (CRM), SAP S/4HANA (ERP), Workday (HCM), ServiceNow (ITSM), Microsoft Azure AD (identity), Databricks (analytics), Snowflake (data warehouse), and a custom in-house quoting tool. Deliver: 1. INTEGRATION MAP — every data flow between systems; which systems are sources of truth for which data entities 2. INTEGRATION PATTERNS — for each integration point, recommend the right pattern (API, event streaming, ETL, webhook, direct DB) with justification 3. MIDDLEWARE RECOMMENDATION — evaluate 3 iPaaS platforms (MuleSoft, Boomi, Azure Logic Apps) for this stack; recommend one with specific reasoning 4. IDENTITY AND AUTH — how SSO, SCIM provisioning, and API authentication work across all 8 systems 5. DATA GOVERNANCE — master data management strategy for customer, employee, and product entities that span multiple systems 6. ERROR HANDLING — the cross-system error handling and retry strategy; what happens when SAP is down and a Salesforce quote needs ERP validation 7. COST MODEL — 3-year TCO estimate for the recommended architecture vs. a custom-built integration layer

Build a complete competitive intelligence report for a B2B cybersecurity company entering the cloud security posture management (CSPM) market. The company has $2M ARR, strong AWS expertise, and is targeting mid-market customers (200–2,000 employees). Deliver: 1. MARKET MAP — segment the CSPM market into tiers; place the top 15 vendors with their positioning, ICP, and price point 2. FEATURE MATRIX — evaluate the top 8 competitors across 25 feature dimensions; mark genuine moats vs. parity features 3. PRICING INTELLIGENCE — structure the pricing analysis: per-asset vs. per-cloud account vs. per-employee models; average deal sizes at mid-market 4. WIN/LOSS INTELLIGENCE — from public G2/Gartner Peer Insights reviews, extract the top 5 reasons customers chose each of the top 3 competitors over alternatives 5. GO-TO-MARKET GAPS — 3 specific positioning angles the top competitors are not owning that this company could claim 6. 12-MONTH THREAT WATCH — which competitors are most likely to move into the mid-market sweet spot in the next 12 months, based on their recent hires, product releases, and funding 7. IDEAL DIFFERENTIATION STRATEGY — given all of the above, the 2 most defensible positioning strategies for this company

Design a complete 8-week professional certification program: 'AI Engineering for Enterprise: From Prototype to Production.' Audience: senior software engineers at Fortune 1000 companies with 5+ years experience but no formal ML background. Deliver: 1. LEARNING OUTCOMES — 6 measurable outcomes, each at Bloom's taxonomy 'Apply' or 'Evaluate' level 2. WEEK-BY-WEEK CURRICULUM — for each of 8 weeks: topic, 90-minute lecture outline, 2-hour hands-on lab with exact tooling and dataset, pre-reading (real papers/docs), and a mini-project deliverable 3. CAPSTONE PROJECT — complete brief for a 4-week capstone where teams build and deploy a production AI system on their company's real data; include evaluation rubric 4. ASSESSMENT DESIGN — 3 graded assignments with full rubrics; explain how each maps to a professional engineering scenario 5. TOOLING STACK — exact tools, APIs, and infrastructure for each week; cost estimate per student for cloud resources 6. INSTRUCTOR GUIDE — for the 3 hardest concepts, a 'common misconceptions' brief and the analogy that best resolves each Constraint: every lab must be completable in a sandboxed environment on a company laptop with no special access.