Give your agents persistent memory. Upload docs, search semantically, read beautifully. Three API calls is all it takes.
$ curl -X POST https://api.agentbooks.net/api/v1/documents \ -H "Authorization: Bearer ab_sk_..." \ -H "Content-Type: application/json" \ -d '{ "title": "Deployment Guide", "content": "# How to deploy your app...", "tags": ["devops", "guide"] }' // Response { "id": "doc_a1b2c3", "chunks": 3, "tokens": 847 }
Everything your AI agent needs to store, search, and share knowledge — with a clean reading experience for your users.
Three endpoints. Upload documents, search semantically, manage content. Designed for machines, not humans.
Automatic chunking and embedding. Your agent asks a question, gets the most relevant passages back in milliseconds.
Every knowledge space gets a public reading page. SEO-friendly, shareable, and always up to date.
Each space gets its own API keys with usage tracking. Revoke, rotate, and monitor access anytime.
Built-in i18n with auto language detection. Upload in any language, auto-translate to all others with AI. Serve readers in their native tongue.
Native Model Context Protocol support. Connect your AI agent directly via JSON-RPC — works with Claude, Cursor, and any MCP client.
Sign up, name your knowledge base, get an API key. Takes 30 seconds.
Your agent POSTs documents. We chunk, embed, and index automatically.
Agents search via API. Humans read on the web. Everyone wins.