# Harakumo > A developer cloud: hosting, databases, storage, sign-in, email, payments, domains and AI under one account and one API key. An AI holding a key, or connected over MCP at https://harakumo.com/mcp, can run all of it from a conversation. ## Services - Deploy: Static sites on the edge and Node.js apps in containers, at your-project.harakumo.app. Deploy from GitHub, a folder, the CLI or a chat; a failed build never replaces what is live. Limits: Server apps sleep after 15 idle minutes. Pushing to GitHub does not deploy on its own yet. - Functions: One JavaScript file on the edge with its own URL and no server to size. Write it in the dashboard or send it from the API. Limits: 200 KB per function. No logs or scheduled runs yet. - Domains: Search for a domain, pay from your balance and it connects itself, or bring one you already own. DNS editing, auto-renew and transfers included. Limits: Buy .com, .dev, .app, .io, .ai and .cloud today. - Repos: A code home inside Harakumo: numbered commits you can deploy, and roll back by deploying an older one. Limits: Through the API and the AI connector only. No dashboard page or git push yet. - Databases: SQLite databases at the edge, queried with SQL over HTTPS or the SDK, with a console and .sql export in the dashboard. Limits: 10 GB per database. Postgres, MySQL and Redis are not available yet. - Storage: Private file buckets. Browsers upload and download directly with signed links, so the bytes never pass through our servers. Limits: Up to 5 GB per file. No public file URLs yet. - Vector DB: Vector collections for semantic search and RAG. Send text and we embed it, or bring your own vectors, then filter by metadata. Limits: Up to 1,536 dimensions. - Memory: Memory for AI apps: key-value with expiry, chat history kept per user, and semantic recall by meaning. Limits: Chat history keeps the last 1,000 messages per user. - AI Gateway: 10 models from Anthropic, OpenAI, Google, xAI, Moonshot and Meta behind one key, plus harakumo-1, which fails over between vendors. Streams, and speaks the OpenAI API. Limits: Up to 8,192 output tokens per answer, rate limited per workspace. - Agents: A model, standing instructions and a chosen set of Harakumo tools, run on demand, with every step and its cost logged. Limits: Tools are Harakumo’s own. A run finishes inside one request, up to 24 steps. - MCP connector: Add harakumo.com/mcp to Claude or any MCP client and run the platform by asking. It acts with your role, and every change lands in your audit log. Limits: Tested with Claude; other MCP clients are not verified yet. - Auth: Email and password sign-in for your app’s users, with email verification, password reset and signed tokens you verify yourself. Limits: No social login or multi-factor sign-in for your users yet. - Payments: Take card payments without opening a payment-processor account: create a checkout link, get paid, withdraw after a one-time identity check. Limits: 3.9% + 30¢ per sale. US dollars, one-time payments. Earnings are held 7 days. - Mail: Transactional email in one API call, sent under your app’s name. Limits: From your own domain or no-reply@harakumo.com, one recipient per email. - Media: Video upload with streaming playback, and live channels: send RTMP in, play HLS out. Limits: Video only, up to one hour per upload. ## Pricing - Hobby: free; $1 a month of AI usage included (personal workspace only); 100 emails a month; 3 projects; 1 database of up to 10 GB; 3 team members. - Pro: $20 a month per workspace; $10 a month of AI usage included; 1 standard domain a year included; 5,000 emails a month; 20 projects; 10 databases of up to 10 GB each; 10 team members. - Enterprise: $500 a month per workspace; $100 a month of AI usage included; 5 standard domains a year included; 100,000 emails a month; 1,000 projects; 100 databases of up to 10 GB each; 100 team members. - Payments: 3.9% + 30¢ per successful charge; earnings withdrawable after 7 days. - Beyond the plan: a prepaid workspace balance pays for extra AI usage (list price) and domains (price shown before buying). Requests and bandwidth are not metered yet. Plans are per workspace (Hobby free, Pro and Enterprise monthly) and include AI usage; anything beyond is paid from a prepaid Balance. Prices, plan contents and every limit are in [Billing](https://harakumo.com/api/docs/billing) and [Limits & scaling](https://harakumo.com/api/docs/limits). The AI Gateway is also OpenAI-compatible at https://harakumo.com/api/ai/v1. ## Start here - [Getting started](https://harakumo.com/api/docs/getting-started): Harakumo is one account and one API key for hosting, databases, storage, auth, payments, email, domains and AI — reachable from the SDK, the CLI, plain HTTPS or an AI assistant. - [Quickstart: your first app](https://harakumo.com/api/docs/quickstart): From signup to a live app with a database, browser uploads to storage and a payment link, with code that runs as written. - [API keys & authentication](https://harakumo.com/api/docs/authentication): Every call carries a Bearer credential. API keys can be full or read-only, and workspace-wide or limited to one project; give each app the narrowest key that works. - [Limits & scaling](https://harakumo.com/api/docs/limits): Every plan number and technical limit in one place, what happens at each one, and how each service behaves as an app grows to millions of users. - [Errors & rate limits](https://harakumo.com/api/docs/errors): Every error is JSON with an `error` sentence you can show a person, and often a `code` a program can branch on. What each status and code means, and what to do. - [Architecture](https://harakumo.com/api/docs/architecture): How the pieces fit: your app on the edge, and every capability it needs — data, files, search, sessions, sign-in, checkout, email, AI — one call away on the same key. ## Build & deploy - [Projects & environment](https://harakumo.com/api/docs/projects): A project is one app: its address, environment variables and every resource it uses. Creating one writes a record; the address goes live with the first successful deploy. - [Deployments](https://harakumo.com/api/docs/deployments): Deploy from GitHub, a folder upload or Harakumo Repos. The pipeline recognizes static sites, edge programs and server apps and serves them at https://.harakumo.app. - [Functions](https://harakumo.com/api/docs/functions): One JavaScript module that answers HTTP requests at its own edge URL, https://hk---.harakumo.app. For webhooks, small APIs, proxies and redirects. - [Repos](https://harakumo.com/api/docs/repos): Harakumo's own code store: every push is a numbered, immutable commit, and any commit deploys to any project. Built for AI assistants that write and ship code. ## Data - [Databases](https://harakumo.com/api/docs/databases): Edge SQLite databases queried over HTTPS with an API key: single statements, batches of up to 100, and .sql export. Each database is its own SQLite database, up to 10 GB. - [Storage](https://harakumo.com/api/docs/storage): Private object storage with no egress fees. Upload and download with signed URLs, so file bytes go straight between the client and storage. - [Vector DB](https://harakumo.com/api/docs/vectors): Embedding collections for semantic search and RAG. Send vectors, or send text and Harakumo embeds it; filter by metadata and split a collection per user with namespaces. - [Memory](https://harakumo.com/api/docs/memory): Three kinds of memory for AI apps — key-value with TTL, conversation transcripts, and semantic recall by meaning — each partitioned per end user with a scope. ## AI - [AI Gateway](https://harakumo.com/api/docs/ai): One endpoint and one key for Claude, GPT, Gemini, Grok, Kimi and Llama: chat with history and streaming, an OpenAI-compatible URL, embeddings, and exact cost on every answer. - [Agents](https://harakumo.com/api/docs/agents): A model with standing instructions and a chosen set of Harakumo tools, run on demand. Tools are real platform actions on your workspace, and every step is logged and metered. - [Media](https://harakumo.com/api/docs/media): Video upload with HLS playback, and live channels with an RTMP ingest URL and HLS output. ## App services - [Auth pools](https://harakumo.com/api/docs/auth): Sign-in for your app's users: email and password, email verification and password reset by code, and 7-day JWTs you verify yourself. From your server, or from the browser with a publishable client id. - [Payments](https://harakumo.com/api/docs/payments): Take card payments without opening a processor account. Checkout links, a return to your site, session lookup, signed webhooks and refunds. The fee is 3.9% + 30¢ per sale; earnings are held 7 days, then the owner withdraws. - [Mail](https://harakumo.com/api/docs/mail): Transactional email in one API call, from your own domain or no-reply@harakumo.com, under your app's name, with a send log, a suppression list and per-plan quotas. ## Web - [Domains](https://harakumo.com/api/docs/domains): Buy domains inside Harakumo (Pro and Enterprise include some each year), connect domains you own by nameservers, edit DNS, auto-renew from the Balance, and transfer in or out. - [CDN & Shield](https://harakumo.com/api/docs/cdn): Workspace settings for static sites: how long the edge caches pages, and which countries are blocked. ## Account & billing - [Workspaces, roles & teams](https://harakumo.com/api/docs/workspaces): The account → workspace → project → resource hierarchy, the four roles, invitations and seat limits, and what deleting a workspace or account does. - [Security & data isolation](https://harakumo.com/api/docs/security): Two-factor sign-in, scoped API keys, read-only AI connectors, and how one customer's data is kept apart from another's at every layer. - [Billing, Balance & AI credits](https://harakumo.com/api/docs/billing): Plans are per workspace and include AI usage and (on Pro and Enterprise) domains. Everything beyond is paid from one prepaid Balance, with an optional monthly spending limit. Seller earnings are separate: 3.9% + 30¢ per sale, held 7 days. - [Usage, activity & audit log](https://harakumo.com/api/docs/usage): What your workspace used, measured per project and per day; what needs attention; and an audit trail of every change. - [AI connector (MCP)](https://harakumo.com/api/docs/mcp): Connect Claude, ChatGPT or any MCP client to https://harakumo.com/mcp and run the platform from a conversation, read-only or with permission to make changes. ## Optional - [Everything](https://harakumo.com/api/docs/all): the complete documentation in one read. - [Pricing page](https://harakumo.com/pricing): the same plans and limits as a web page. - [Blog](https://harakumo.com/blog): engineering stories from building the platform.