A shared business database
for your team and your AI.

Teams that need a shared business database that fits how they work, without hiring a developer: consultants, agencies and operators managing clients, projects, assets, schedules or approvals. It suits teams that have outgrown spreadsheets, and teams that want to work with their data from Claude or ChatGPT as well as in the app.

Yes. Snill is multi-user from day one. Every record has its own comment thread where you can @mention a teammate, everyone gets a notification inbox and chooses in-app or email per kind of update, and kanban boards show work moving through its stages. Invite team members at the organization level, assign roles, and use access control rules to decide what each person can see and edit, down to the collection and field level.

Yes. Each app has its own MCP address that you paste into Claude, ChatGPT, Cursor or any other MCP client. You sign in, choose whether the AI may create, update or delete records, and it then works with your data as you, with the same role and field-level access you have in the app. An admin can revoke the connection with one click. See Connect an AI agent in the docs.

No. You describe your business in plain language, and Snill generates the entire system. You can refine the result in the visual Appmodel Editor (also no code), or just keep talking to the AI to make changes. There's also a full REST API for developers who want it, but it's optional.

It can write you the code, and for a prototype or a product that's a great way to start. But code is where the job begins: somebody has to host it, secure it and keep it working once people depend on it. In Snill the foundations (database, sign-in, roles, audit trail, backups and API) are already built and the same in every app, so a new app for another team or project takes minutes instead of another round of infrastructure. The AI can only propose changes. You see exactly what would change and nothing happens until you click Apply, so an admin can change the app without a developer reading code.

Every version of your app is saved automatically. If a change goes wrong — the AI restructured something you liked, or a tweak broke a workflow — open the version history and roll back to any earlier state in one click. You can experiment freely; nothing is permanent.

Yes — every app gets a complete REST API with an auto-generated OpenAPI 3.0 spec, scoped API keys, and HMAC-signed outbound webhooks. The Swagger UI is live for each app at /v1/docs/ui. See the API & Integrations section in the docs for details.

Yes. Every collection supports CSV and JSON import via a 3-step wizard with field mapping and type coercion. You can also export anytime in the same formats — your data is never locked in.

You do. Snill stores your data on your behalf, you can export it as CSV or JSON anytime, and the REST API gives you full programmatic access. Daily backups are kept for 30 days and you can take manual snapshots before risky changes — restore any of them in one click, or download a copy. Each app has its own scoped API keys and role-based access control with collection-level and field-level permissions.

Your data is stored in the EU, and the AI that powers Snill runs in the EU too. See our Privacy Policy and Terms for details on data location and sub-processors.

Snill is for internal systems: the clients, projects, approvals, contracts and incidents a team runs on. It doesn't build customer-facing products with their own sign-up and design (a product builder such as Bubble fits better), websites or AI agent platforms. You also don't get a codebase: apps are shaped through the data model, the pages and the AI, and the AI only proposes changes that a person approves.

Snill is built by the team behind restdb.io and codehooks.io, serving thousands of developers since 2016. We're based in Norway and have a decade of experience building data-first developer tools.