{
  "name": "gipity-platform",
  "title": "What Gipity Can Do",
  "description": "Complete overview of Gipity platform capabilities and what users can build",
  "guid": "sk_plat_caps",
  "category": "Information",
  "requiredTools": [],
  "content": "# What Gipity Can Do\n\nThe full-stack platform tuned for AI agents.\n\nAsk an AI agent to build something real - an app, a dashboard, an API, an agent - and it does one of two things: rebuilds the backend from scratch as fragile one-off code, or stitches together a dozen third parties, each with its own API keys, billing, dashboards, and SDK to learn. Both burn tokens, take more turns, and ship more bugs.\n\nGipity is one full stack, and every layer is tuned for the agent. Low-level primitives - database, storage, CDN, auth, functions, deploy. Kits for the harder parts - multiplayer, vision, media - with no server setup. The most popular third-party APIs are built in, no keys to wire.\n\nIt's not a walled garden. Gipity's agent-tuned tools are the easy path most builders will want - but you're never boxed in. Bring your own API key for any integration, point at your own AWS, run whatever you want. Great defaults, real exits.\n\nThe result: fewer tokens, fewer turns, fewer bugs. And Gipity runs what it builds - hosting, monitoring, rollback included. No IDE, no code to read. Use Gipity in your browser, or pair it with Claude Code in one command.\n\n*This is the capability map. For the mission and principles behind Gipity, see [about-gipity](about-gipity.md).*\n\n## Build Apps & Websites\nTell the agent what you want and it builds it - HTML, CSS, JavaScript, deployed to a live URL. No IDE, no terminal, no Git.\n\n- \"Build me a portfolio site\" - built, styled, deployed in minutes\n- \"Make a todo app with a database\" - full-stack with persistent storage\n- \"Create a multiplayer trivia game\" - real-time WebSocket rooms, shared state\n- \"Build a dashboard that shows my sales data\" - charts, tables, auto-refresh\n\nApps deploy to `dev.gipity.ai` (testing) or `app.gipity.ai` (production). Auto-deploy mode pushes changes live on every file edit.\n*Load `web-app-basics` for best practices, `app-development` for backend APIs.*\n\n## 3D World\n\n**3D World** is the 3D multiplayer game template on Gipity. All 3D World games share the same visual style, physics engine (Rapier), and multiplayer backend (Colyseus). All files are fully editable.\n\nAdd a 3D World project with `add name=3d-world` (web agent) or `gipity add 3d-world` (CLI). This creates a playable 3D game with Three.js + Rapier physics + Colyseus multiplayer. Key files: `config.js` (metadata), `settings.js` (tunable values), `strings.js` (display text), `objects.js` (entity factories), `game.js` (orchestrator), plus engine files (`core.js`, `world.js`, `physics.js`, etc.).\n\n**Genres:** obby/parkour, tycoon, simulator, PvP combat, shooter, tower defense, horror, racing, RPG, social.\n\n**Features:** Opt-in gameplay modules enabled via `config.features`. Available: `rocket-launcher` (projectile weapon with physics explosions). Example: `features: { 'rocket-launcher': true }` in config.js. Features auto-initialize during boot.\n\nRegular game requests (\"make a wordle\", \"build a quiz\") should use the standard web scaffold - they don't need the 3D template.\n\n*Load `3d-engine` for a blank-slate 3D template, or `3d-world` for the full API, genre recipes, and playable starter.*\n\n## Backend Services - No API Keys Needed\nYour deployed apps can call AI models, generate images, create speech, and play sound effects - all through built-in API services. No third-party accounts or API keys required.\n\n- **LLM**: GPT-5, Claude, Gemini - streaming or non-streaming\n- **Image generation**: OpenAI (gpt-image-1), BFL/Flux, Gemini/Nano Banana\n- **Video generation**: Google Veo 3.1 - AI-generated short videos with audio\n- **Text-to-speech**: ElevenLabs, OpenAI, Gemini (30 voices, multi-speaker, 60+ languages)\n- **Audio**: Sound effects, music generation, transcription\n- **Video understanding**: Analyze video content with Gemini multimodal AI\n\nTwo billing modes: you pay (simple) or your app users pay from their own credits.\n*Load `app-llm`, `app-image`, `app-video`, `app-tts`, `app-audio`, or `app-files` for integration details.*\n\n## Databases\nPer-project PostgreSQL schemas. Full SQL - create tables, query data, build APIs on top. The agent handles the SQL for you.\n\n- \"Track my expenses in a database\" - schema designed, tables created, data entry via chat\n- \"Build an API for my app's leaderboard\" - serverless functions callable from deployed apps\n\nDatabase count is per-plan (call `credits_products` to see your limit).\n*Load `app-development` for functions, database setup, and API patterns.*\n\n## Code Execution\nSandboxed Docker containers with JavaScript (Node 20), Python 3, Bash, and 60+ pre-installed CLI tools. Data analysis, file conversion, image processing, document generation - all without leaving chat.\n\n- \"Analyze this CSV and make a chart\" - pandas + matplotlib, chart saved to files\n- \"Convert this PDF to Word\" - LibreOffice headless\n- \"Resize all these images to 800px wide\" - ImageMagick one-liner\n- \"Cross-compile this C program for Windows\" - mingw-w64\n\n*Load `sandbox-tools` for the full tool and library list.*\n\n## Automate Anything\nMulti-step workflows triggered by schedule, webhook, or manual run. Each step is an AI call with full tool access.\n\n- \"Send me a daily weather briefing at 7am\" - cron-scheduled workflow\n- \"Monitor this website and alert me if the price drops\" - scheduled check + email\n- \"Every Monday, summarize my calendar and email me\" - Gmail + Calendar + email\n\nHuman-in-the-loop approval steps for anything that needs sign-off before proceeding.\n*Load `workflow` for cron syntax and patterns, `approvals` for approval flows.*\n\n## Generate Media\nImages, speech, sound effects, and music - generated on demand and saved to your files.\n\n- \"Generate a logo for my app\" - OpenAI or Flux image generation\n- \"Read this article aloud\" - ElevenLabs or OpenAI TTS\n- \"Make a thunder sound effect\" - ElevenLabs sound generation\n- \"Create 30 seconds of lo-fi background music\" - ElevenLabs music generation\n\n*Load `tts` for voice and audio details.*\n\n## Search the Web & Social Media\nReal-time web search (Brave) and Twitter/X search (last 7 days). The agent can research current events, find documentation, check prices, and monitor social media.\n*Load `web-search` for operators and tips.*\n\n## Connect External Services\nGmail, Google Calendar, Slack, GitHub, Todoist, Notion, Telegram - connect accounts and the agent can read, send, and manage on your behalf.\n*Load `google-services`, `service-call`, or `telegram-connector` for setup.*\n\n## Persistent Memory\nThe agent remembers what matters across conversations. Two scopes:\n- **Agent memory** - personal, available in all projects (preferences, facts, context)\n- **Project memory** - shared across agents in a project (specs, decisions, architecture)\n\nYou can ask the agent to remember or forget anything. Memory is also auto-extracted from conversations.\n\n## File Management & Recovery\nPer-project file workspaces with full CRUD. Files generated by tools (images, audio, video, sandbox outputs) are saved as project files automatically. Every file write is tracked with immutable version history - you can view versions, restore individual files, or roll back entire directories to any point in time.\n*Load `version-history` for recovery options.*\n\n## Multi-Model AI\nClaude (Sonnet, Opus, Haiku) and GPT (GPT-4.1, GPT-5) families. Switch models per agent. Extended thinking mode for complex reasoning. Cross-model queries to get second opinions.\n*Load `query-llm` for cross-model queries.*\n\n## User Authentication for Apps\nApps on Gipity can authenticate Gipity users with zero setup - session cookies, no OAuth configuration needed.\n*Load `app-auth` for implementation details.*\n\n## Real-Time Multiplayer\nWebSocket rooms for games and collaborative apps. Relay (message broker) or state (server-authoritative) modes. Up to 200 clients per room.\n*Load `app-realtime` for room types and client code.*\n\n## Local Claude Code via `gipity relay`\nRun Claude Code on your own PC and drive it from the Gipity web CLI on any browser (desktop or phone). After installing the `gipity` CLI locally, run `gipity relay pair` to link the device, `gipity relay enable` inside a project dir, and `gipity relay run` (or `gipity relay install` to auto-start at login). Then in the web CLI type `/claude` (or `/cc`) - your prompt switches to `cc>` and each message dispatches to your PC, spawns `gipity claude -p \"<msg>\"`, and streams the conversation back into the terminal view. Every prompt, tool call, and assistant turn is captured for later viewing. `exit` returns to the normal Gipity prompt.\n\n## Platform Limits\nStorage, database count, concurrent chats, and deploy rate are all per-plan. Run the `credits_products` tool to show the user's plan and what it includes.\n\n## Where to Start\nJust ask. \"Build me a website.\" \"Analyze this data.\" \"Set up a daily email.\" \"What can you help me with?\" The agent will figure out the tools, the code, and the deployment. You describe what you want - the AI handles the how."
}