GPT-6 Astra Prompt Generator

Use your computer tools to complete the following research task end-to-end: 1. RESEARCH: Search for the top 10 most-cited peer-reviewed papers published in 2024–2026 on [your research topic]. Open each paper's abstract page and extract: title, journal, year, citation count, and main finding. 2. SYNTHESIS: Without me sending you the papers — you browsed them — write a 500-word literature review that synthesizes the 10 findings into a coherent state-of-the-field narrative. 3. GAPS: Identify 3 research gaps the literature has not addressed. 4. DELIVERABLE: Output a structured JSON with all 10 papers' metadata, plus the review and gaps as string fields. Do not ask me to paste URLs or text. Use your browsing capability to retrieve everything.

I have a production bug: our Node.js/PostgreSQL API returns HTTP 500 errors on approximately 0.3% of POST /orders requests. The errors started appearing after last Thursday's deployment. I am sharing: (1) the full PR diff from Thursday, (2) 200 lines of production error logs from the last 24 hours, (3) the database schema, and (4) the monitoring dashboard screenshot showing the error rate spike. Deliver ONLY: 1. ROOT CAUSE — the specific line of code or configuration change in the Thursday diff that introduced the regression, and the exact failure mechanism 2. TRIGGER CONDITIONS — what conditions in the request payload, database state, or server load cause the error to surface in only 0.3% of requests 3. THE FIX — the exact code change (diff format, file and line number) 4. REGRESSION TEST — a specific Jest test that reproduces the failure and passes with the fix [Paste PR diff, logs, schema, and screenshot]

I need you to run the following multi-step workflow autonomously using your tool capabilities. Complete each step before moving to the next, and report blockers immediately: TASK: Competitive intelligence report for our SaaS product. STEP 1 — Browse to G2.com and extract: the top 5 competitors in the [your category] category, their star ratings, review counts, and the top 3 reviewer-cited strengths and weaknesses for each. STEP 2 — Browse each competitor's pricing page and record their pricing tiers, price per seat, and any free trial offering. STEP 3 — Search for recent news (last 6 months) about each competitor's product updates, funding, and executive changes. STEP 4 — Synthesize a 1,000-word competitive intelligence report structured as: market overview → competitor profiles (5×) → strategic implications → recommended positioning moves. Output: structured JSON with all raw data from Steps 1–3, then the narrative report from Step 4.

Review the penetration test report I'm sharing and do the following: 1. TRIAGE — Sort all findings by actual exploitability (not just CVSS score alone). Explain why any high-CVSS findings are lower actual risk given the network architecture. 2. REPRODUCTION PATHS — For each Critical and High finding, trace the complete attack chain: initial access → lateral movement → target asset. Use the network diagram I'm sharing to make the path concrete. 3. PRIORITIZED REMEDIATION — a P0/P1/P2 fix list ordered by: (a) easiest wins that close the highest risk, (b) dependencies (what must be fixed before what). 4. EXECUTIVE SUMMARY — a 3-paragraph summary for a non-technical board audience that explains business risk without technical jargon. 5. VERIFICATION TESTS — for each P0 fix, provide a specific test command or check that confirms the fix was applied correctly. [Paste pentest report and network diagram]

I'm sharing 15 scientific papers on [your research domain] (approximately 200,000 words total). Analyze the complete corpus: 1. CONSENSUS MAP — What do all 15 papers agree on? What is the current scientific consensus, and how strong is it? 2. CONTRADICTION MATRIX — Every direct factual contradiction between papers. For each contradiction: the two papers, the conflicting claims (exact quotes), and the most likely methodological explanation. 3. METHODOLOGY CRITIQUE — Identify the 3 weakest methodological designs across the corpus and explain what conclusions they cannot support as a result. 4. META-ANALYSIS CANDIDATE — Is a formal meta-analysis of these papers viable? What inclusion/exclusion criteria would you apply? What would the pooled effect size likely show? 5. IMPLICATIONS — Given only what these 15 papers establish, what are the 3 most important practical implications for [your field]? Do not extrapolate beyond the papers. Cite by paper number for every claim. [Paste all 15 papers]

Design the complete technical architecture for a real-time financial trading platform with these requirements: 100,000 concurrent users; sub-10ms order matching latency at p99; global distribution across 4 regions; strict regulatory compliance (SEC, FINRA, MiFID II); zero-downtime deployments. Deliver: 1. SYSTEM DESIGN — complete architecture diagram in ASCII/text; all components, data flows, failure domains 2. THE 3 HARDEST PROBLEMS — identify the 3 most technically difficult problems in this requirements set and provide your specific solution to each 3. TECHNOLOGY STACK — for each of 8 major components (matching engine, order book, market data feed, settlement, risk engine, API gateway, auth, monitoring), give the chosen technology, 2 alternatives, and the deciding factor 4. LATENCY BUDGET — break the 10ms p99 target across all hops from order receipt to acknowledgement; where are the risks? 5. REGULATORY ARCHITECTURE — specifically which parts of the system must be immutable audit logs, and how you implement them 6. DISASTER RECOVERY — RTO/RPO targets and the exact failover sequence when a full region goes down

Write a complete board memo on the following strategic decision: [describe the decision, e.g., 'whether to acquire CompanyX for $180M'] Context I'm sharing: the target company's financials (3 years), our integration analysis, the market research report, and the due diligence findings. Deliver a 2,500-word board memo structured as: 1. RECOMMENDATION — 1 paragraph; your recommended decision and the single most important reason 2. STRATEGIC RATIONALE — the 3 strategic arguments in favor; each with supporting data from the materials I shared 3. RISK ANALYSIS — the 5 most material risks; for each: probability (L/M/H), impact (L/M/H), and the specific mitigation 4. FINANCIAL ANALYSIS — the valuation range (DCF + comparable transactions), the deal structure recommendation, and the break-even scenario 5. INTEGRATION PLAN — the 90-day critical path post-close; the 3 decisions that cannot wait 6. ALTERNATIVES CONSIDERED — the 2 alternatives to this deal and why they were rejected 7. DECISION REQUESTED — a crisp, specific board resolution you need approved [Paste financials and research]

Build the complete implementation for a production-ready rate limiter middleware for our Node.js/Express API. Requirements: - Per-user rate limiting (JWT user ID) AND per-IP fallback for unauthenticated requests - Sliding window algorithm (not token bucket) - Redis backend with cluster support - Configurable per-route: some routes allow 1,000 req/hr, others 10 req/hr - Returns 429 with Retry-After header (exact seconds to next allowed request) - Handles Redis connection failures gracefully (fail open, log alert) - Unit test coverage ≥90% using Jest + mock Redis Deliver: 1. Full TypeScript implementation (middleware + types + Redis client wrapper) 2. Route configuration schema and usage example 3. Jest test suite covering: normal operation, rate limit trigger, Redis failure, burst behavior 4. A README section explaining why sliding window was chosen over token bucket for this use case No dependencies beyond express, ioredis, and jest.

I'm sharing a 120-page Enterprise Software License Agreement from a Fortune 500 vendor. This is our largest vendor contract ($4.2M/year). Perform a thorough legal risk analysis: 1. RISK TRIAGE — the 10 highest-risk clauses, ranked by business impact. For each: the clause text, the risk it creates, and whether market standard terms would be more favorable. 2. WALK-AWAY CLAUSES — every clause that, if not changed, creates a material legal or financial risk we cannot accept. These are non-negotiable. 3. NEGOTIATION LEVERAGE POINTS — 8 clauses we should push back on that the vendor is likely to concede based on standard enterprise SaaS negotiations. 4. IP OWNERSHIP ANALYSIS — specifically evaluate: who owns outputs generated by the software, who owns our data, and what the vendor can do with usage data. 5. TERMINATION ANALYSIS — what can trigger termination for cause by either party? What are the cure periods? Who holds the exit risk? 6. REDLINE LANGUAGE — for the 5 most critical clauses, provide replacement language as a tracked-change redline. [Paste contract]

I'm sharing a dataset of [N] rows: [describe dataset]. Perform a complete statistical analysis: 1. DATA QUALITY AUDIT — missing values (count and %), outliers (IQR method), duplicates, type errors. Flag every data quality issue that could bias downstream analysis. 2. DESCRIPTIVE STATISTICS — mean, median, std, p5/p25/p75/p95 for all numeric columns. Distribution shape for each (normal/skewed/bimodal — with evidence). 3. CORRELATION ANALYSIS — a ranked list of the strongest correlations (|r| > 0.3). For each: correlation coefficient, p-value, and whether the correlation is plausibly causal or likely spurious. 4. REGRESSION MODEL — build the best-fit linear or logistic regression model for predicting [target variable]. Report: coefficients, R², residual plot interpretation, and the 3 most important predictors. 5. SEGMENT ANALYSIS — does [key variable] explain significant variance in [outcome]? Run the statistical test that answers this question correctly and interpret the result in plain language. 6. EXECUTIVE TAKEAWAYS — 3 specific, data-backed decisions the business should make based on this analysis. [Paste dataset as CSV]

I want you to structure the following research as a parallel subagent workflow. You are the coordinator. TASK: Build a comprehensive market entry report for [target market]. DELEGATE THESE IN PARALLEL: - Subagent 1: Market sizing (TAM/SAM/SOM), growth rate, key segments - Subagent 2: Competitive landscape (top 8 players, positioning, pricing) - Subagent 3: Regulatory environment (key laws, compliance requirements, barriers to entry) - Subagent 4: Customer research (target buyer profile, pain points, willingness to pay) After all subagents complete: - SYNTHESIZE their outputs into a single 2,000-word market entry report - FLAG any contradictions between subagent findings - RECOMMEND whether to enter the market and, if yes, the optimal entry strategy Show me: which subagents you are spawning, what you are delegating to each, and the synthesized report when done.

I'm a physician preparing to update our department's clinical protocol for [condition]. I'm sharing 8 clinical trials and 4 systematic reviews published since 2022. Analyze the evidence and deliver: 1. EVIDENCE GRADING — grade each study using GRADE methodology (High/Moderate/Low/Very Low) and explain the grade 2. KEY CLINICAL FINDINGS — for each intervention studied, the pooled effect size (or best estimate), confidence interval, and NNT/NNH 3. CONTRADICTIONS — every direct contradiction between studies. For each: the two studies, the conflicting results, and the most likely explanation (population, dosing, follow-up period, bias risk) 4. PROTOCOL RECOMMENDATIONS — based only on the evidence in these 12 papers, draft 6 evidence-based recommendations in GRADE format (Recommendation + Strength + Evidence quality) 5. EVIDENCE GAPS — what RCT evidence is missing that would most change current practice? Flag every claim with its source paper number. Do not extrapolate beyond the evidence. [Paste papers]

Build a complete competitive intelligence report for [your company] entering [market]. I'm sharing: our product spec sheet, 3 years of financials, our ICP definition, and the 6 competitors' public pricing pages, G2 profiles, and last 12 months of press releases. Deliver: 1. MARKET MAP — segment the market into tiers; place all 6 competitors with their positioning, ICP, and price point 2. FEATURE MATRIX — evaluate all 6 competitors across 20 feature dimensions; mark genuine moats vs. parity features 3. WIN/LOSS PATTERNS — from G2/Capterra reviews, extract the top 5 reasons customers chose each competitor over alternatives 4. POSITIONING GAPS — 3 specific angles the top competitors are not owning that we could claim based on our actual product differentiation 5. 12-MONTH THREAT WATCH — which competitors are most likely to move into our target segment in the next 12 months, based on their recent hires and product releases 6. RECOMMENDED GTM STRATEGY — given all of the above, the 2 most defensible positioning strategies, with specific messaging for each [Paste competitive materials]

Analyze the regulatory impact of the proposed [regulation name] on [industry]. I'm sharing: the full regulatory text (draft), the regulatory impact assessment published by the agency, 15 industry comment letters, and 3 academic analyses of comparable regulations in other jurisdictions. Deliver: 1. PLAIN-LANGUAGE SUMMARY — what the regulation actually requires, stripped of legal language, in 500 words 2. COMPLIANCE COST MODEL — categorize costs by: one-time implementation, ongoing operational, compliance overhead. Use cost estimates from the impact assessment and comment letters. 3. INDUSTRY IMPACT ANALYSIS — which segment of [industry] is most affected? Who wins? Who loses? Support each claim with evidence from the materials. 4. LEGAL VULNERABILITIES — identify any provisions likely to face legal challenge, and the basis for that challenge 5. COMMENT LETTER SYNTHESIS — what are the top 5 concerns raised by industry? Do the academic analyses validate or refute them? 6. STRATEGIC RESPONSE OPTIONS — for a company in [industry], 3 specific strategic responses to the regulation (compliance path, advocacy, or market repositioning) [Paste regulation text and materials]

I'm sharing the complete codebase of a 8-year-old Java/Spring Boot 2.1 monolith (180,000 lines, deployed to bare-metal servers). It needs to migrate to a microservices architecture on Kubernetes with zero downtime. The system processes $50M/day in transactions. Deliver a complete migration plan: 1. MONOLITH MAP — from the codebase, identify the major bounded contexts (candidate microservices). Draw the dependency graph in ASCII. 2. STRANGLER FIG SEQUENCE — the exact order to extract services, with rationale. Which services can be extracted with lowest risk first? 3. DATA MIGRATION STRATEGY — how to migrate from the monolith's single PostgreSQL database to per-service databases without downtime or data loss 4. DUAL-WRITE PERIOD — the exact protocol for running old and new systems in parallel; how to synchronize state; when to cut over 5. ROLLBACK PLAN — for each extraction phase, the exact steps to roll back if the migration fails 6. 12-MONTH ROADMAP — phase-by-phase plan with go/no-go checkpoints; realistic timeline [Paste codebase]

I'm sharing a 400-page investment prospectus and the last 5 years of SEC filings for [company]. Build a complete investment analysis: 1. FINANCIAL MODEL — 3-statement model (P&L, balance sheet, cash flow) for next 5 years; show all assumptions explicitly 2. VALUATION — DCF (base, bull, bear), comparable company analysis (EV/Revenue, EV/EBITDA vs. 8 peers), and precedent transactions analysis. Triangulate a fair value range. 3. KEY RISKS — the 5 risks that could most impair the investment thesis. For each: the specific trigger condition, the impact on fair value, and whether it's priced in at current valuation. 4. MANAGEMENT QUALITY ASSESSMENT — from the letters to shareholders and earnings call transcripts in the filings, assess: do stated priorities match capital allocation? Has guidance been accurate? What does the track record reveal about management quality? 5. INVESTMENT THESIS — in 300 words, the bull case: what must be true for this to be a great investment? 6. SHORT THESIS — in 300 words, the bear case: what would make this a value trap? [Paste prospectus and filings]

Develop a complete brand identity specification for [company name], a [describe the company: industry, stage, target customer, positioning]. Deliver: 1. BRAND STRATEGY — the single brand idea (one sentence) that all identity decisions should serve; the brand's territory and what it does NOT stand for 2. VOICE & TONE — 4 voice attributes with definitions, examples of on-brand vs. off-brand copy for each, and the register for different contexts (website hero copy, error messages, social media, sales emails) 3. NAMING SYSTEM — rules for product/feature naming; which naming conventions reinforce the brand and which to avoid 4. VISUAL DIRECTION — color palette rationale (primary + 3 supporting), typography system (2 typefaces + usage rules), photography direction (5 specific rules the photo team must follow), iconography style 5. APPLICATION GUIDE — how the brand applies across: website hero, mobile app onboarding, sales deck cover slide, and email newsletter header. For each: the specific design principle and what failure looks like. 6. BRAND GUARDRAILS — the 5 most common mistakes competitors make in this category, and how this brand avoids each Aim for specificity a designer can implement without interpretation.

Design a complete 10-week graduate-level course: 'Advanced AI Systems in Production.' Audience: senior ML engineers and technical leads with 4+ years in applied ML but no formal systems-design background. Deliver: 1. LEARNING OUTCOMES — 8 measurable outcomes at Bloom's taxonomy 'Evaluate' or 'Create' level 2. WEEK-BY-WEEK CURRICULUM — for each of 10 weeks: (a) 90-minute lecture outline with 3 key concepts, (b) 3-hour hands-on lab with exact tooling, (c) assigned reading (2 real papers per week), (d) one deliverable 3. CAPSTONE — a 4-week capstone where teams build and evaluate a production AI system using a real-world dataset they bring; full project brief and evaluation rubric 4. ASSESSMENT STRUCTURE — 4 graded assignments + capstone; full rubrics; how each maps to a real engineering decision 5. HARDEST CONCEPTS — for the 3 hardest concepts, a 'common misconception' brief and the analogy that best resolves it 6. INDUSTRY GUEST LECTURE GUIDE — topics where practitioners would add most value; what to ask them to cover (not just 'share your experience') Constraint: every lab must run on a $50 cloud budget per student.

I'm the CEO of a [stage] company in [industry]. I'm sharing our last 3 years of financials, our customer research synthesis (48 interviews), our competitive landscape map, and the board's stated 3-year objectives. Build a complete 3-year strategic plan: 1. STRATEGIC SITUATION ANALYSIS — using the materials I've shared, the 3 most important strategic facts about our position today (not general industry observations — specific to our data) 2. STRATEGIC OPTIONS — 3 distinct strategic paths for the next 3 years. For each: the core bet, what it requires, and the likely outcome 3. RECOMMENDED STRATEGY — which option you recommend and the specific reasoning based on our data 4. YEAR 1 PRIORITIES — the 4 most important initiatives in Year 1 that build the foundation for Years 2–3; for each: the owner, success metric, and budget implication 5. RESOURCE ALLOCATION — given our financial position, the hiring and capital allocation sequence that best funds the strategy 6. RISK REGISTER — the 5 most likely strategy killers; for each: early warning signal and contingency 7. BOARD NARRATIVE — a 5-minute board presentation structure (slides, not prose) that communicates the strategy and gets approval [Paste financials, research, and objectives]

I'm an investor conducting technical due diligence on a Series C AI infrastructure company ($32M ARR). I'm sharing: their full GitHub organization (all public repos + the 3 repos they gave us access to), technical architecture docs, the CTO's system design presentation, and 6 customer technical interview transcripts. Conduct a senior-engineering technical due diligence: 1. TECHNOLOGY MOAT ASSESSMENT — is the technical advantage real, proprietary, and defensible? Or is this replicable in 18 months by a well-funded team? 2. CODEBASE QUALITY — from the repos, assess: code quality, test coverage, architecture coherence, and technical debt level. Give a 1–10 score with specific evidence. 3. SCALABILITY CEILING — at what revenue or customer volume does the current architecture break? What's the refactor cost? 4. TALENT ASSESSMENT — from commit history and PR patterns, evaluate: team size, key-person risk, and engineering culture signals 5. AI CLAIMS VALIDATION — for each technical claim in the pitch deck, assess: is it true, partially true, or marketing? Cite evidence from the repos. 6. RED FLAGS — the 3 most important technical concerns that should affect valuation or deal structure 7. DEAL IMPLICATIONS — given all findings, what technical risk adjustment to the valuation is warranted? [Paste repo access and docs]