Monday, June 22, 2026

I want to build an app called "SavedSpot". ## The Problem People constantly save Instagram Reels and TikToks about restaurants, events, and pop-ups but forget about them because they get buried in saves with no reminder or location-aware surfacing. ## Target Audience Urban dwellers and food-and-event enthusiasts who save social media posts as wish lists but consistently forget about them. ## Core Idea Resurfaces your saved Instagram and TikTok food, event, and pop-up posts as location-aware reminders at exactly the right moment. People constantly save restaurant recommendations and event posts on Instagram and TikTok, but those saves get buried and forgotten because the platforms provide no reminders or location context. SavedSpot lets users import their saved posts, automatically extracts the venue or event name using AI, geocodes it, and sends a push notification when the user is physically nearby or when an event date is approaching. It works across both platforms through their public share URLs with no API key required from the user. ## Monetization Strategy Free for up to 20 saved spots; $3/month for unlimited spots, cross-platform sync, and calendar integration for event reminders. ## Requirements - Category: Productivity - Difficulty: Week - Suggested stack: Next.js + localStorage or Supabase + PWA Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "ArtPoison". ## The Problem Artists don't want their work used to train LLMs but find current poisoning tools like Glaze technically complex and slow to apply image-by-image, as discussed in the Lobsters thread about LLM poisoning of artwork. ## Target Audience Independent artists, illustrators, and photographers who publish work online ## Core Idea Automatically apply imperceptible adversarial perturbations to artwork images before publishing, making them resistant to LLM and diffusion model training. Artists and illustrators want to share their work online without contributing to AI training datasets, but manually applying glaze or nightshade-style protection to every image is tedious and requires technical knowledge. ArtPoison provides a simple drag-and-drop web app or browser extension that batch-processes artwork with state-of-the-art poisoning techniques before upload. It targets artists on Tumblr, DeviantArt, and personal sites who want protection without managing command-line tools. ## Monetization Strategy Freemium — free for up to 20 images/month, $8/month for unlimited batch processing and priority GPU queue ## Requirements - Category: AI/ML - Difficulty: Week - Suggested stack: Next.js + Anthropic Claude API + Vercel AI SDK Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "BuddyRevive". ## The Problem Claude Code's /buddy companion feature was silently removed with no changelog entry, devastating thousands of developers who had formed genuine attachment to it, as evidenced by the 2,019-upvote GitHub issue. ## Target Audience Claude Code users who miss the companion feature and solo developers who benefit from morale-boosting interactions during long coding sessions ## Core Idea A Claude Code extension that restores the /buddy companion experience with a customizable persona, mood-aware status line, and session continuity across restarts. Anthropic silently removed the /buddy feature from Claude Code v2.1.97 with no changelog entry, generating one of the most emotionally charged GitHub issues in the repo with 2,019 upvotes and 262 comments from developers who had formed genuine attachment to the companion. BuddyRevive is a Claude Code plugin that intercepts the /buddy slash command and implements a local companion layer with configurable name, personality, and motivational messages that persist across sessions. It stores state in a local SQLite file so the buddy remembers context from previous coding sessions. ## Monetization Strategy Free and open-source to drive adoption; optional $4/month cloud tier for syncing buddy memory across machines ## Requirements - Category: Developer Tool - Difficulty: Weekend - Suggested stack: Node.js CLI or VS Code extension + TypeScript Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "AlertPin". ## The Problem County-wide emergency alerts go to millions of people regardless of proximity to the actual hazard, leaving residents unable to determine their personal risk level without navigating complex GIS sources. ## Target Audience Homeowners and renters in fire-prone, flood-prone, or severe-weather regions who receive vague county-level emergency alerts ## Core Idea Paste your address and instantly see whether it falls inside the precise risk polygon for any active emergency alert in your county — not just whether your county is under the alert. Emergency alerts like Red Flag Warnings, flood watches, and evacuation orders are broadcast county-wide to hundreds of thousands of residents, but the actual hazard polygon is often a small fraction of the county. Residents currently have no easy way to check whether their specific address is inside the real risk zone without navigating multiple government GIS layers. AlertPin geocodes a user-supplied address, fetches NWS alert polygons in real time, and renders a clear yes/no with a map showing their pin relative to the exact hazard boundary. ## Monetization Strategy Free for single address lookups; $2.99/month for saved addresses with push notifications when any saved location enters or exits an active alert polygon ## Requirements - Category: Health - Difficulty: Weekend - Suggested stack: Next.js + Supabase + PWA + Chart.js Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "NotionRescue". ## The Problem Notion iOS users report deleted workspaces with no recovery, voice-to-text broken for months, and crashing comments, but feel unable to migrate years of accumulated notes without a reliable export tool. ## Target Audience Heavy Notion users who store critical personal or professional data and are anxious about losing it to app bugs with no recovery option. ## Core Idea One-click Notion workspace backup and export to Markdown, with automatic daily snapshots so you never lose years of work to a silent deletion bug. Notion's iOS app has deleted entire workspaces with no recovery path, and Notion support has told users they 'can't find' the lost data. NotionRescue runs as a background service that exports your entire Notion workspace daily to local Markdown files with preserved hierarchy, links, and attachments. It also provides a one-click restore preview so you can verify your backup is complete before anything goes wrong. ## Monetization Strategy $5/month for automated daily backups; $10/month for hourly snapshots and cross-device restore; one-time $29 lifetime plan for individuals. ## Requirements - Category: Productivity - Difficulty: Week - Suggested stack: Next.js + localStorage or Supabase + PWA Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "BuddyBack". ## The Problem Claude Code's /buddy companion feature was silently removed with no changelog entry, devastating thousands of developers who had formed genuine attachment to it, as evidenced by the 2,020-upvote GitHub issue. ## Target Audience Claude Code users and AI-assisted developers who relied on /buddy for morale and focus during long coding sessions. ## Core Idea A drop-in Claude Code companion that restores the /buddy experience with a local personality layer you fully control. When Anthropic silently removed /buddy from Claude Code v2.1.97, over 2,000 developers filed one of the most emotionally charged GitHub issues in the repo's history. BuddyBack is a lightweight CLI wrapper that intercepts Claude Code sessions and injects a persistent companion persona using a local config file, giving developers the morale-boosting presence they lost. It runs entirely on the developer's machine with no cloud dependency and supports custom personality profiles. ## Monetization Strategy Free core with a $4/month Pro tier for multi-persona profiles, team-shared configs, and session mood analytics. ## Requirements - Category: Developer Tool - Difficulty: Weekend - Suggested stack: Node.js CLI or VS Code extension + TypeScript Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "WorkVis". ## The Problem Corporate SWEs report that real engineering contributions are invisible to management while performative busy work gets rewarded, causing top performers to be undervalued with no automated way to surface their actual output. ## Target Audience Software engineers at mid-to-large companies who feel their contributions are undervalued and want objective evidence for performance reviews. ## Core Idea Automatically surfaces your real engineering impact to management so your actual output — not your visibility — determines how you're evaluated. Corporate engineers at large companies report that performative busy work gets rewarded while genuine technical contributions remain invisible to management. WorkVis connects to git, Jira, and CI systems to generate weekly impact digests that translate commits, PR reviews, and incident responses into plain-English narratives tailored for non-technical managers. It runs locally and never sends code to the cloud, only metadata summaries. ## Monetization Strategy $8/month per seat, with a free tier for individual contributors and a $30/month team plan for shared dashboards. ## Requirements - Category: Productivity - Difficulty: Week - Suggested stack: Next.js + localStorage or Supabase + PWA Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "ScienceMode". ## The Problem Researchers doing legitimate biology and immunology work find mainstream AI tools like Claude excessively censor routine scientific questions, making them nearly unusable for research workflows. ## Target Audience Academic researchers, PhD students, and biotech professionals who need AI assistance for standard laboratory science without fighting over-cautious safety filters. ## Core Idea An AI research assistant tuned for biology, chemistry, and immunology that doesn't refuse routine scientific questions that mainstream tools over-censor. Researchers doing legitimate biology and immunology work find that mainstream AI tools like Claude and ChatGPT excessively refuse routine scientific questions about pathogens, reagents, and experimental protocols, making them nearly unusable for daily lab research. ScienceMode wraps a local or privacy-focused model with domain-specific system prompts and a curated scientific knowledge base that answers standard research questions without false-positive safety refusals. It includes citation lookups, protocol templates, and reagent calculation tools built specifically for wet-lab and computational biology workflows. ## Monetization Strategy $19/month for individual researchers; $49/month per lab seat for team plans with shared protocol libraries; annual academic discounts. ## Requirements - Category: AI/ML - Difficulty: Month - Suggested stack: Next.js + Anthropic Claude API + Vercel AI SDK Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "EmbedAI". ## The Problem Embedded engineers can't use generic AI coding tools because they hallucinate register addresses, generate code for peripherals that don't exist on the target chip, and confuse quirks between similar MCU variants, causing hard-to-debug hardware failures. ## Target Audience Embedded systems engineers, firmware developers, and hardware hackers working with microcontrollers who need AI assistance that respects hardware constraints. ## Core Idea An AI coding assistant that knows your exact microcontroller — registers, peripherals, and errata — so it never hallucinates hardware details for your chip. Generic AI coding tools hallucinate register addresses, generate code for peripherals that don't exist on the target chip, and confuse quirks between similar MCU variants, causing hard-to-debug hardware failures in embedded projects. EmbedAI lets engineers upload their chip's datasheet, SVD file, and errata document, then answers hardware questions and generates peripheral initialization code grounded strictly in those documents. It cross-references every generated register access against the SVD before returning a response, flagging any ambiguity rather than guessing. ## Monetization Strategy $15/month for individual engineers with up to 5 chip profiles; $40/month for team plans with shared chip libraries and private datasheet uploads. ## Requirements - Category: Developer Tool - Difficulty: Month - Suggested stack: Node.js CLI or VS Code extension + TypeScript Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "MailEmbedKit". ## The Problem SaaS companies need embeddable email editors for their users but existing solutions have prohibitively expensive licensing costs that make them inaccessible for bootstrapped or early-stage products. ## Target Audience Indie hackers and small SaaS teams building products that need to let their end users compose or customize HTML emails ## Core Idea An affordable, embeddable drag-and-drop email editor component for SaaS products — priced for bootstrapped teams, not enterprises. SaaS companies that want to let their users compose or customize HTML emails need an embeddable editor, but existing solutions like Stripo, Unlayer, and Beefree charge enterprise pricing that is prohibitive for bootstrapped or early-stage products. MailEmbedKit is a self-contained React and Vue component that provides a full drag-and-drop email editor with template management, merge tag support, and HTML export, distributed under a startup-friendly flat-fee license. It specifically targets indie hackers and small SaaS teams building features like email campaign composers, transactional template editors, or newsletter builders. ## Monetization Strategy $49/month flat for startups under $10K MRR, $199/month for growth stage, with a perpetual license option at $999 for self-hosted deployments ## Requirements - Category: SaaS - Difficulty: Month - Suggested stack: Next.js + Supabase + Stripe Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "BlogRoll". ## The Problem Developers seeking well-designed text-heavy personal blogs for inspiration or reading cannot find them because search is dominated by agency portfolios and marketing listicles rather than genuine individual writing. ## Target Audience Developers and writers looking for reading-focused blog design inspiration or high-quality personal technical blogs to follow ## Core Idea A curated, searchable directory of genuinely text-focused personal and developer blogs — no agency portfolios, no marketing listicles, just real writing. Developers redesigning personal blogs or looking for reading-focused web design inspiration cannot find curated examples of well-designed text-heavy sites because every search result is dominated by agency showcases and portfolio sites. BlogRoll is a human-curated and community-nominated directory where every submission is reviewed for being a genuine individual writing-focused blog, with tags for design style, topic focus, and post frequency. The Lobsters developer tools thread and tech minimalism thread both reflect a hunger for authentic, non-commercial technical writing communities. ## Monetization Strategy Free to browse; $5/month for submitters who want a featured listing badge and analytics on referral traffic from the directory ## Requirements - Category: Social - Difficulty: Weekend - Suggested stack: Next.js + Supabase Realtime + Auth Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "ActionsAllow". ## The Problem GitHub Actions matrix jobs have no native allow-failure support per individual job and no multi-choice input type, forcing teams into brittle workarounds that break status checks and create confusing CI output. ## Target Audience DevOps engineers and platform teams using GitHub Actions for complex CI/CD pipelines with matrix builds and manual deployment triggers. ## Core Idea Adds native allow-failure and multi-choice input support to GitHub Actions without touching your YAML workflow files. GitHub Actions has no native allow-failure support for individual matrix jobs and no multi-choice input type for manual workflows, forcing teams into complex shell-in-YAML workarounds that break status checks and create confusing CI output. ActionsAllow is a GitHub App that post-processes workflow run results and a companion schema extension that unlocks multi-select inputs via a clean UI overlay. Teams install it once and immediately get cleaner CI without rewriting their workflow files. ## Monetization Strategy Free for public repositories; $8/month per private organization with unlimited repositories. ## Requirements - Category: Developer Tool - Difficulty: Week - Suggested stack: Node.js CLI or VS Code extension + TypeScript Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "ContribPulse". ## The Problem Open-source contribution graphs gamify over-commitment, and maintainers have no tool to detect when contributors are heading toward burnout before they burn out or disappear, as raised in a 1,789-upvote GitHub issue. ## Target Audience Open-source project maintainers and community managers who care about sustaining contributor health and retention. ## Core Idea Detects open-source contributor burnout signals before they disappear, giving maintainers a heads-up so they can intervene early. The GitHub contribution graph gamifies over-commitment, and maintainers have no tool to detect when contributors are heading toward burnout before they burn out or silently disappear. ContribPulse analyzes commit cadence, PR response times, and tone shifts across issues and comments to generate a burnout-risk score for each contributor. Maintainers receive a weekly digest highlighting at-risk contributors so they can proactively reach out, reduce scope, or redistribute load. ## Monetization Strategy $15/month per repository, free for repositories with fewer than 20 contributors; discounts for non-profit and FOSS foundations. ## Requirements - Category: SaaS - Difficulty: Week - Suggested stack: Next.js + Supabase + Stripe Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "MoodStream". ## The Problem Users want a music app with a single-tap play button and vibe customization by activity and mood, rather than having to select a specific playlist — a gap identified in the Software Recommendations request. ## Target Audience Casual music listeners who find playlist management tedious and want context-aware instant playback ## Core Idea A music player that lets you set your current activity, mood, and energy level and instantly starts a fitting playlist — no browsing, no curation required. Music apps require users to pick a playlist before they can start listening, creating friction when someone just wants music that fits their current vibe. Inspired by Yandex Music's one-tap play and vibe features, MoodStream offers a single large Play button that starts music immediately based on three sliders: activity, mood, and energy. It connects to Spotify or Apple Music as a backend and learns preferences over time, targeting users who find playlist management exhausting. ## Monetization Strategy $3.99 one-time purchase on iOS and Android, with optional $1.99/month for cross-device sync and listening history ## Requirements - Category: Productivity - Difficulty: Week - Suggested stack: Next.js + localStorage or Supabase + PWA Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "LeanCheck". ## The Problem Technical founders build products without validating assumptions, and fifteen years after The Lean Startup there is still no lightweight tool that enforces customer discovery discipline and tracks hypothesis invalidation across interviews. ## Target Audience Early-stage founders, indie hackers, and product teams who want to validate ideas before building. ## Core Idea A structured customer discovery tool that forces founders to track hypotheses, run experiments, and prove demand before writing a single line of code. Fifteen years after The Lean Startup, Eric Ries himself acknowledged in his HN AMA that founders still build without validating assumptions first. LeanCheck provides a lightweight framework for logging customer discovery interviews, tracking which hypotheses have been invalidated, and generating a living assumption map that shows what you still don't know. It replaces the informal sticky-note chaos of early-stage validation with an auditable, shareable record. ## Monetization Strategy Freemium with up to 3 active experiments free; $12/month for unlimited experiments, team collaboration, and PDF export for investor sharing. ## Requirements - Category: SaaS - Difficulty: Week - Suggested stack: Next.js + Supabase + Stripe Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "LockWidget". ## The Problem Users are frustrated by having to fully unlock their phone every time they need to view or add a quick note, especially during activities like grocery shopping — a pain point flagged in the previously generated ideas list as LockNote. ## Target Audience iOS users who frequently reference short lists or notes during hands-busy activities ## Core Idea A lock-screen widget app for iOS that lets you read and append quick notes without unlocking your phone. People engaged in activities like grocery shopping, cooking, or working out need to glance at or add a note without the friction of unlocking their phone. iOS 16+ lock-screen widgets allow interactive experiences but no existing notes app has built a proper lock-screen note widget with append capability. LockWidget provides a minimal, secure lock-screen widget that displays your current list and lets you tap to add items using a numeric PIN rather than Face ID or full unlock. ## Monetization Strategy $1.99 one-time purchase on the App Store with a free tier limited to one note ## Requirements - Category: Productivity - Difficulty: Weekend - Suggested stack: Next.js + localStorage or Supabase + PWA Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "ForgeFlow". ## The Problem Jujutsu and other non-Git VCS users have no forge that supports change-centric workflows — GitHub and GitLab assume branch-based PRs that map poorly to how these tools actually work, as discussed in the Lobsters forge feature request thread. ## Target Audience Developers using Jujutsu, Sapling, Pijul, or Darcs who need a code review and collaboration platform that matches their VCS model ## Core Idea A forge built for Jujutsu and change-centric version control workflows, where the unit of review is a change rather than a branch. GitHub and GitLab assume branch-based pull requests that map poorly to Jujutsu's change-centric model, forcing JJ users into awkward workarounds or abandoning the forge entirely. ForgeFlow is a self-hostable forge where the primitive is a change rather than a branch — reviews are attached to change IDs, stacked changes are visualized natively, and rebasing or amending a change automatically updates open reviews. The 91-comment Lobsters thread shows active demand from Jujutsu, Sapling, and Pijul users who feel homeless in the current forge ecosystem. ## Monetization Strategy Open-core: free self-hosted tier, $15/user/month for cloud-hosted with CI integration and team management ## Requirements - Category: Developer Tool - Difficulty: Month - Suggested stack: Node.js CLI or VS Code extension + TypeScript Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "DiceAtlas". ## The Problem Tabletop gamers and game designers cannot calculate or visualize probability distributions for complex custom dice formulas involving conditional expressions, as no accessible tool handles anything beyond simple additive rolls. ## Target Audience Tabletop RPG designers, board game developers, and players who need to analyze complex custom dice mechanics ## Core Idea Visualize full probability distributions for any complex dice formula, including conditional expressions and reroll mechanics, directly in your browser. Tabletop game designers and players need to understand probability distributions for custom dice formulas like 'minimum of 3d6 times the sum of 2d10, reroll ones' but existing tools only handle simple additive expressions. DiceAtlas parses a rich dice formula language supporting min, max, reroll, exploding dice, and conditional expressions, then renders the full probability distribution as an interactive chart. It targets game designers validating balance, players optimizing character builds, and educators teaching probability. ## Monetization Strategy Free web tool with a $5 one-time purchase for the desktop app with offline support and formula library export ## Requirements - Category: E-commerce - Difficulty: Week - Suggested stack: Next.js + Shopify API or Stripe Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "MultiGitID". ## The Problem Developers working with both personal and work GitHub accounts must manually switch credentials constantly, with no tool that automatically applies the right identity per repository. ## Target Audience Software developers who maintain separate personal and work git identities and are frustrated by constant manual credential switching. ## Core Idea Automatically switches your git identity — name, email, SSH key, and signing key — based on which repository you're working in. Developers working with both personal and work GitHub accounts must manually update git credentials every time they switch repositories, leading to commits accidentally attributed to the wrong identity. MultiGitID installs a global git hook that reads a simple per-directory config file and applies the correct identity transparently with no manual steps. It supports SSH key switching, GPG signing keys, and GitHub Desktop integration, addressing a 1,348-upvote GitHub issue. ## Monetization Strategy One-time purchase of $9 for the GUI configuration app; CLI core remains open source. ## Requirements - Category: Developer Tool - Difficulty: Weekend - Suggested stack: Node.js CLI or VS Code extension + TypeScript Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.

I want to build an app called "ChinaSort". ## The Problem GNU Coreutils sort cannot order CJK numbers by semantic value regardless of locale options, leaving developers with no scriptable tool for sorting Chinese numerals or text by meaning rather than Unicode code point. ## Target Audience Developers and data engineers processing Chinese, Japanese, or Korean text in scripts or pipelines ## Core Idea A tiny CLI and library that correctly sorts CJK text by semantic numeric value, radical order, or pinyin — something GNU coreutils simply cannot do. Developers and data engineers working with Chinese, Japanese, or Korean text data have no reliable command-line tool for sorting CJK content semantically. GNU coreutils sort fails on CJK numerals regardless of locale flags, leaving users with no scriptable alternative. ChinaSort is a small cross-platform CLI (and importable library) that parses Unicode kPrimaryNumeric fields, pinyin romanization, and stroke-count metadata to sort CJK strings correctly. ## Monetization Strategy Open-source CLI with a hosted REST API tier at $5/month for teams needing high-volume sorting without self-hosting ## Requirements - Category: Developer Tool - Difficulty: Weekend - Suggested stack: Node.js CLI or VS Code extension + TypeScript Please help me build this step by step. Start with: 1. A project structure and initial setup 2. The core data models 3. The main feature implementation 4. A simple but polished UI Keep it lean — MVP first, ship fast. Use modern best practices and make it production-ready.