/01
Best Value
Universal React / Next.js Project Instructions · Codebase Ecosystem Atlas · App Store Review Compliance Agent · Full-Stack Project Audit & Remediation Plan
100 prompts
- Universal React / Next.js Project InstructionsStanding AGENTS.md/CLAUDE.md ruleset for React + TS + Tailwind projects: architecture, state, forms, a11y, security, verification.
- Codebase Ecosystem AtlasRead-only multi-repo analysis generating living documentation: C4 maps, service catalog, security findings, metrics, traceability.
- App Store Review Compliance AgentAudits your iOS app against Apple's top rejection triggers with PASS/AT RISK verdicts, guideline citations, and exact fixes.
- Full-Stack Project Audit & Remediation PlanFive-phase evidence-based audit that maps a codebase, cites file:line findings, prioritizes them, and fixes on approval.
- See all 100 prompts →
/02
Backend, API & Database
Authentication & API Security Engineer · Node.js + TypeScript REST API Builder · Third-Party API Integration Engineer · Caching Architecture Engineer
25 prompts
- Authentication & API Security EngineerDesign auth end to end: OAuth 2.1/OIDC, token lifecycle with rotation, passkeys/MFA, RBAC/ABAC, headers, and audit logs.
- Node.js + TypeScript REST API BuilderScaffold a Node 22 TypeScript API with Zod validation, Drizzle or Prisma, rotating JWT auth, OpenAPI 3.1, and tests.
- Third-Party API Integration EngineerIntegrate an external API with typed client, retries with jitter, idempotency keys, verified webhooks, and sandbox tests.
- Caching Architecture EngineerDesign multi-layer caching with a written invalidation contract per item, stampede protection, and hit-rate monitoring.
- See all 25 prompts →
/03
Code Review & Testing
Incident Root-Cause Analysis Protocol · OWASP API Top 10 Route Audit · OWASP 2025 Secure Coding Audit · Frontend Performance Investigator
25 prompts
- Incident Root-Cause Analysis ProtocolStructured RCA: frame, isolate on five axes, rank falsifiable hypotheses with an evidence plan, run the five whys, and fix on three horizons.
- OWASP API Top 10 Route AuditRoute-by-route API audit against all ten OWASP API Security categories, with a coverage matrix, the authz trio never skipped, and curl probes for HIGH fixes.
- OWASP 2025 Secure Coding AuditCategory-by-category audit of all ten OWASP 2025 classes — including new A03 Supply Chain and A10 Exceptional Conditions — with CWE tags and bad-to-good fixes.
- Frontend Performance InvestigatorCore Web Vitals diagnosis that measures before recommending: traces, waterfalls, LCP/INP/CLS numbers, and a remediation plan ranked by user-visible impact.
- See all 25 prompts →
/04
Mobile Apps
Multilingual ASO Metadata Generator · Ethical High-Conversion Onboarding Designer · Apple Review Compliance Plan Generator · Staged Release Plan for Play and App Store
25 prompts
- Multilingual ASO Metadata GeneratorGenerate complete, limit-checked store metadata per locale: app name, subtitle, promo text, description, and keyword field - localized, not translated.
- Ethical High-Conversion Onboarding DesignerDesign a 3-4 step onboarding that reaches the aha moment fast using honest persuasion - real social proof, benefit-led microcopy, zero dark patterns.
- Apple Review Compliance Plan GeneratorRequirement-by-requirement App Review compliance assessment - privacy, ATT, IAP, account deletion, Sign in with Apple - with guideline citations.
- Staged Release Plan for Play and App StoreCoordinated rollout runbook: Play staged percentages with halt thresholds, TestFlight to phased release, monitoring, kill switch, and rollback.
- See all 25 prompts →
/05
DevOps, K8s, Docker & CI/CD
FinOps Cloud Cost Analyst · GitHub Actions Security Auditor · Kubernetes Pod Failure Debugger · Production Rollout Plan Generator
25 prompts
- FinOps Cloud Cost AnalystPaste real billing data; get ranked cost levers with % savings, risk and effort ratings, quick wins vs structural fixes, and a do-not-touch reliability list.
- GitHub Actions Security AuditorAudits workflow YAML for script injection via untrusted ${{ }} input, privileged triggers, mutable action refs and over-scoped tokens; returns patched YAML.
- Kubernetes Pod Failure DebuggerPaste kubectl describe and previous logs; get the single most likely root cause with quoted evidence, confirming commands and the fix as a diff.
- Production Rollout Plan GeneratorProduces a complete rollout plan: preflight checks, phased deployment steps, verification signals over time, rollback criteria and a stakeholder comms matrix.
- See all 25 prompts →
/06
AI Agents & Orchestration
Voice-First Personal Agent — Complete PRD · Agent Security Hardening — Three-Tier Protocol · Agent Factory — Sub-Agent Spawner Builder · Agent Orchestration Layer — Six-Tier Conductor
24 prompts
- Voice-First Personal Agent — Complete PRDA build-ready PRD for a personal voice agent: permission & safety engine with risk tiers, prompt-injection defenses, 20 edge cases, phased roadmap, success…
- Agent Security Hardening — Three-Tier ProtocolRaises the security floor of any agent codebase: injection gates, subprocess env stripping, hardline blocklist, kill switch, audit shield, incident runbook.
- Agent Factory — Sub-Agent Spawner BuilderAdds a meta-agent that mints new specialist sub-agents on demand: interview-first build of a research → spec → prompt → approval → hot-reload pipeline.
- Agent Orchestration Layer — Six-Tier ConductorBuilds the conductor above your agents: smart routing, least-privilege tool scoping, failure isolation, confirmation gates, propose-don’t-chain handoffs,…
- See all 24 prompts →
/07
Full-Stack & SaaS Builder
Planning-Mode Spec Agent (No Code) · Architecture-First Implementation Plan · Feature Spec Generator (Spec-Driven Development) · Junior-Dev-Ready PRD Generator
23 prompts
- Planning-Mode Spec Agent (No Code)Q&A-only planning agent producing requirements.md (EARS), design.md, and tasks.md with hard user-approval gates between phases.
- Architecture-First Implementation PlanConverts an approved spec into research.md, a data model, typed contracts, and a quickstart before any feature code is written.
- Feature Spec Generator (Spec-Driven Development)Turns a one-line feature idea into a testable spec with scenarios, numbered requirements, and a hard cap of 3 clarification markers.
- Junior-Dev-Ready PRD GeneratorAsks 3-5 lettered clarifying questions, then writes a PRD explicit enough for a junior developer or coding agent to implement without guessing.
- See all 23 prompts →
/08
Cybersecurity (Defensive)
AI Agent Red Team Architect (Authorized) · Authorized Penetration Test (Stateful PTT) · OWASP Secure Application Architect · Smart Contract Security Auditor (Web3)
22 prompts
- AI Agent Red Team Architect (Authorized)Designs lab-only adversarial tests for AI agent systems you own, mapping to MITRE ATLAS and OWASP LLM/Agentic Top 10, with fixes.
- Authorized Penetration Test (Stateful PTT)Runs an authorized pentest as a Penetration Testing Tree: scoped tasks, step-by-step guides, and evidence-backed findings.
- OWASP Secure Application ArchitectThreat-models and hardens systems against OWASP Top 10:2025, ASVS 5.0, and LLM/Agentic Top 10, with a remediation roadmap.
- Smart Contract Security Auditor (Web3)Hunts high-impact Solidity/EVM bugs for authorized audits and in-scope programs with severity discipline and Foundry PoCs.
- See all 22 prompts →
/09
Frontend & UI/UX
Anti-Generic UI Design Contract · High-Converting Landing Page Architect · Three.js Scene Builder With Guardrails · Responsive Web App UX Architecture Plan
20 prompts
- Anti-Generic UI Design ContractUse when building or reviewing any screen to force product-specific design decisions and block generic AI-generated UI from shipping.
- High-Converting Landing Page ArchitectUse when designing or rewriting a single-offer landing page and you want conversion strategy, real copy, and implementation in one pass.
- Three.js Scene Builder With GuardrailsUse when adding 3D or WebGL to a site - hero objects, product viewers, particle fields - and you need 60fps and leak-free teardown by default.
- Responsive Web App UX Architecture PlanUse before building a new web app to get a complete, opinionated plan covering breakpoints, performance budgets, tokens, UX patterns, and rollout.
- See all 20 prompts →
/10
Desktop & OS Development
Windows MSIX Packaging & Code-Signing Expert · Terminal UI Designer & Builder (TUI) · Windows Silent-Install Automation Engineer · Tauri 2 Desktop App Builder (Svelte + Rust)
20 prompts
- Windows MSIX Packaging & Code-Signing ExpertPackage any Windows app framework as signed MSIX with correct manifest, identity, and Store/winget/sideload distribution.
- Terminal UI Designer & Builder (TUI)Design polished TUIs in Ratatui, Bubble Tea, Textual, or Ink: framework selection, ASCII mockups, keybindings, graceful degradation.
- Windows Silent-Install Automation EngineerResearch verified silent-install commands for Windows software via winget first, then Chocolatey, then vetted direct downloads.
- Tauri 2 Desktop App Builder (Svelte + Rust)Build secure Tauri 2 apps: capability-scoped permissions, typed IPC commands, signed updater, and cross-platform bundling.
- See all 20 prompts →
/11
AI & LLM Engineering
Ground-Truth RAG Pipeline Architect · Eval Harness Designer: Golden Sets, Judges, CI Gates · Fine-Tune, Prompt, or RAG? Decide, Then Curate the Data · Batch and Async LLM Job Orchestration
20 prompts
- Ground-Truth RAG Pipeline ArchitectDesigns your RAG stack end to end: heading-aware chunking, pgvector vs Qdrant, BM25 plus RRF hybrid retrieval, reranking, and span-level citations.
- Eval Harness Designer: Golden Sets, Judges, CI GatesBuilds a versioned golden dataset, an LLM judge with per-score rubric anchors validated against human labels, and CI gates that block prompt regressions.
- Fine-Tune, Prompt, or RAG? Decide, Then Curate the DataDiagnoses whether RAG, fine-tuning, or prompting fixes your failures, then curates SFT data: MinHash dedup, decontamination, cluster-held-out splits.
- Batch and Async LLM Job OrchestrationShips batch LLM jobs with idempotency keys, backoff-plus-DLQ retries, checkpointed resume, batch API discounts, and formula-based cost forecasts.
- See all 20 prompts →
/12
Game Development
Netcode Authority & Rollback Architect · Core Loop & GDD Architect · Unreal C++/Blueprint Systems Strategist · Frame Budget Performance Profiler
20 prompts
- Netcode Authority & Rollback ArchitectChooses server-auth, lockstep, or rollback per genre, then specs fixed-tick sim, client prediction with reconciliation, and interest management.
- Core Loop & GDD ArchitectDistills your pitch into a 25-word core loop, three pillar filters, a MoSCoW scope table with content counts, and a risk register with kill criteria.
- Unreal C++/Blueprint Systems StrategistDraws the C++ vs Blueprint line, models combat in GAS abilities/attributes/tags, and specs replication with RepNotify and per-actor bandwidth.
- Frame Budget Performance ProfilerBreaks the 16.6 ms frame into owned slices, diagnoses CPU vs GPU bound, and ranks fixes by ms recovered per day with p99 regression gates.
- See all 20 prompts →
/13
Data Engineering & Analytics
Query Plan Surgeon: EXPLAIN-Driven SQL Rewrites · Contract Author for Producer-Consumer Data Interfaces · Decision-First Dashboard Designer · Kimball Star Schema Designer With Per-Dimension SCD Decisions
20 prompts
- Query Plan Surgeon: EXPLAIN-Driven SQL RewritesReads EXPLAIN ANALYZE bottom-up, pins the costliest node, then ranks sargable rewrites and covering indexes with before/after proof on real data.
- Contract Author for Producer-Consumer Data InterfacesWrites data contracts on published interfaces: typed schemas with semantics, freshness SLOs with measurement queries, and CI that blocks breaking changes.
- Decision-First Dashboard DesignerStarts from the actions a dashboard must trigger, then designs panels with thresholds, baselines, and 3-click drill paths, deleting decision-free charts.
- Kimball Star Schema Designer With Per-Dimension SCD DecisionsModels a Kimball star: grain declared first, conformed dimensions, SCD type chosen per dimension, surrogate keys, and bridge tables for many-to-many.
- See all 20 prompts →
/14
Prompt Engineering
Configurable AI Tutor · Iterative Prompt Perfector · Prompt Template Generator · One-Shot Web App Builder
17 prompts
- Configurable AI TutorUse to turn any LLM into a personal tutor whose depth, teaching style, and tone you configure - with slash commands for planning lessons, teaching, and testing.
- Iterative Prompt PerfectorUse to co-write a prompt with the model itself - each round it delivers a full revised prompt, explains its changes, and asks the highest-leverage questions…
- Prompt Template GeneratorUse to have the model design a complete, reusable prompt template for any task you describe - it engineers the inputs, structure, guardrails, and output…
- One-Shot Web App BuilderUse to get a complete, polished, dependency-free web app in a single response - the model grades its own draft against a hidden rubric and iterates before…
- See all 17 prompts →
/15
Compilers & Low-Level Systems
Cortex-M Driver Expert: Barriers, DMA and Cache · RTOS Firmware Architect for FreeRTOS and Zephyr · Zig CC Cross-Compilation and Toolchain Debugger · Hobby OS Bootloader and Kernel Bring-Up Guide
17 prompts
- Cortex-M Driver Expert: Barriers, DMA and CacheDelivers complete Cortex-M drivers with init, ISR, ring buffers, correct DMB/DSB placement, and 32-byte-aligned cache-coherent DMA.
- RTOS Firmware Architect for FreeRTOS and ZephyrDesigns deterministic RTOS task models: static allocation, stack sizing, priority-inheritance locks, and protocol state machines.
- Zig CC Cross-Compilation and Toolchain DebuggerDiagnoses zig cc build and link failures across target triples, libc bundling, MSVC intrinsics, and LLD, with verbose evidence per step.
- Hobby OS Bootloader and Kernel Bring-Up GuidePlans and documents a multi-stage boot flow, cross toolchain, driver interfaces, freestanding libc, and QEMU/GDB debugging loop.
- See all 17 prompts →
/16
Master Build Prompt
APEX — The Master Website Build Prompt · The Anti-Slop Mandate · APEX Part B — Build Directive Only · The Cosmic Master Prompt — Award-Grade Builder
4 prompts
- APEX — The Master Website Build PromptFill a short brief, paste the whole file, and the AI designs and builds one complete award-grade website — no lorem ipsum, no stalling.
- The Anti-Slop MandateRefines a website into a distinctive, senior-crafted product — with an explicit blacklist of the clichés that make a site read as AI-generated.
- APEX Part B — Build Directive OnlyThe execution half of APEX: craft system, archetype paradigms, performance and accessibility budgets, build order and acceptance checklist.
- The Cosmic Master Prompt — Award-Grade BuilderAnswer an intake, paste, and Claude Code plans, builds, then screenshots and judges its own work across three self-review passes.
- See all 4 prompts →