Introduction
What Klarety is, who it is for and how these docs are organised.
Klarety is one place where a team brings its data in, asks it questions, builds AI agents and apps on it, and puts those apps online. Files, tables, code and AI tools share one , so there is nothing to wire together first.
What you can do with Klarety
- Keep your data in one database. Upload a file, turn a spreadsheet into a table or bring data in from a tool you use. Tables live in , your database, and you query them in from the app or from code.
- Build in a Studio. Every has a : a cloud workspace with your code, a terminal, a live preview and AI that can change your project. Use Klarety's AI or your own Claude Code, Codex or Cursor subscription.
- Let agents do recurring work. Describe a task in plain words, such as "summarise yesterday's orders every morning", and an runs it on a schedule, when data arrives or when you press Run. It asks before changing anything important.
- Put an app online. Deploy to a free address or your own domain, and roll back with one click.
- Work as a team. Invite teammates, decide who can see which data, and share projects.
Who it is for
These docs serve two kinds of reader:
- People who work in the app. Analysts, operators and founders who want answers and working tools without much code. Everything can be done from the app.
- Developers. Engineers who reach the same data and features through the Python , the
klaretycommand line tool and the HTTP API.
The "For everyone" and "For devs" switch
Many tasks can be done two ways. The switch in the top bar picks your view:
- For everyone shows the buttons and pages you click.
- For devs shows the same task with the Python SDK, the command line or the API.
This browser remembers your choice. Pages that only make sense one way, such as the API reference, show the same content in both views.
You are reading the app view. Steps name the page to open, such as SettingsMembers, and the button to press.
You are reading the developer view. Examples use the Python SDK:
from klarety import Nexus
nx = Nexus.connect("YOUR_API_KEY")
print(nx.sql("SELECT 1").rows)How these docs are organised
The sidebar follows the order most people meet Klarety in:
| Section | What it covers |
|---|---|
| Get started | This introduction, the Quickstart, Core concepts and Security and trust. |
| Guides | End-to-end walkthroughs that combine several features, such as your first project or a daily report. |
| Projects | Creating projects, working in a Studio, project settings, secrets and people. |
| AI harness | The AI in every Studio: Klarety's AI, Claude Code, Codex and Cursor, copies and the Klarety MCP. |
| Data | Nexus tables, the SQL editor, object storage, the knowledge graph, data access and restoring data. |
| Agents | Creating agents, templates, runs and taking an agent live. |
| Deploy | Putting an app online, run settings, approvals, logs and domains. |
| Connectors | Bringing data in from GitHub, Google Drive, PostgreSQL, Amazon S3 and other tools. |
| Workspace | Members, roles, billing, credits and settings. |
| Developers | The Python SDK, the command line tool and the API reference. |
Have a goal in mind? Start with a guide. Need the detail of one feature? Go to its section.
Where to start
- New to Klarety? Follow the Quickstart.: from sign-up to your first query.
- Want the vocabulary first? Read Core concepts for what a workspace, project, Studio and Nexus are.
- Evaluating Klarety for your team? Read Security and trust.
- Writing code? Switch to For devs and start with the Developers section.