Why
Ask. Don't buy.
Software at the speed of a question.
autario is the infrastructure underneath your data apps. It pulls your data together, keeps it accessible in one shape, and runs the apps you put on top. Integration, refresh, hosting and maintenance are taken care of.
What this is for
Data layer
Public data and your own data, in one shape
Vetted public statistics from the big publishers, next to your own tools, connected once and refreshed automatically, or uploaded. One country code, one time column, one unit.
See the data layer
App layer
Prompt to production. You build, we run.
Describe the app your team needs. Your AI builds it through MCP in five calls, on data that is already there, and it goes live on a URL your team opens.
See how an app ships
Where you work
The same numbers, where you already work
Pull the same data into Power BI, Excel or your LLM over the public API and MCP. One read path, one set of numbers, no export step in between.
Read the API docs
Freedom
Freedom to operate. Freedom to be creative.
No purchasing process per tool. No specialty subscription for the fraction of it you use. The customizations are yours, and they change when the question changes.
See what a seat covers
Use cases
What people run on it today
Brands watch how AI assistants talk about them. Marketing teams read search, analytics and ads in one place. Analysts build the app their team was missing.
See the use cases
What this is for
Most apps come down to two things: have data, and do something with the data. Almost all the time and money goes into the first one. Figures pasted into a deck by hand, a subscription bought for the one report you read a quarter, a purchasing process that outlasts the analysis, the same cleaning done again every month, and a week of waiting for a number that already exists.
autario is the layer underneath. It pulls your data together, keeps it accessible in one shape, and runs the business and data apps you put on top of it. Integration, refresh, hosting and maintenance are handled here.
The one thing that can still ask something of you is a provider changing its own API or its own schema. When that happens we detect the change, write it to a changelog your app can read, and the rework is a small one instead of a rebuild.
What is left is the part worth paying attention to: expensive specialty software you no longer need, a first version the same day instead of after a quarter of demos, and full ownership of both layers. Freedom to operate. Freedom to be creative.
Vetted public data, not hallucinated
World Bank, Eurostat, OECD, WHO, IMF, FRED, the ECB, the US Census and the SEC. Every number keeps the link to the publisher that issued it, and every table arrives in the same shape: one country code, one time column, one unit. That is what turns a join across two publishers into a filter. Nothing is generated, and a number that cannot be traced does not ship.
Explore the data
Your own tools, connected once
Analytics, search, ads, commerce, social and your AI spend. We run the OAuth apps, the token refresh and the schema changes, so connecting is a click and yesterday is still there tomorrow. Anything else joins as a REST connector with a key you paste once, or as a file you upload.
Connect your own
Prompt to production
A data app is a page that answers one question on live data. Your AI builds it through MCP, in five calls, and nobody opens a canvas.
The five calls
- create_app names it and opens a workspace.
- update_app_manifest declares the data it is allowed to read.
- write_app_artifact or upload_app_bundle carries the code.
- get_app_preview_url opens it privately, before anyone else sees it.
- publish_app puts it on a URL your team opens.
Built through MCP in five calls, published unlisted.
Then read the same numbers back out wherever you already work, in an LLM, in Power BI or in Excel, over the public API and MCP. Connect via MCP
Do you feel the business vibes?
Read the developer docs
Where you work
The data layer has one read path, and it does not care what opens it.
Power BI and Excel
Point them at the REST endpoint and refresh from there. Filters travel with the request, so a report pulls the rows it needs instead of a whole table.
How to read data over REST
Your LLM
Connect the MCP endpoint and ask in words. Claude, ChatGPT, Gemini or anything else that speaks MCP, with no provider wired in.
The MCP tool reference
Your own code
The same endpoints, one key, and a quickstart that gets a first app running end to end.
Start the quickstart
What people run on it today
Four things this layer already carries. Every one of them runs on data that is connected once and stays current.
Brands and marketing agencies
Track how AI assistants answer questions about your brand: who gets named, in what tone, and which sources the assistants cite. Bring your own model key and the running cost is the tokens.
AI Visibility 360
Social media managers
Connect Instagram, Facebook, TikTok, YouTube and LinkedIn once. Every channel then writes into the same daily shape, so comparing them is a filter and not a spreadsheet.
See the social connectors
Marketing teams
Search Console, Google Analytics 4 and your ad accounts land in one place on one schedule, so a weekly report reads from one table instead of four exports.
See the marketing connectors
Analysts and developers
Build the app your team is missing on public and connected data, preview it privately and publish it on a URL you share. The app pictured above was built exactly that way.
Build one yourself
Ask for the answer, not the table
The same question twice. On the left in words, on the right the way the publisher's own API wants it. Both sides are measured, not estimated.
With autario
- Ask
- What was the unemployment rate in Germany from 2010 to 2020?
query_dataset(dataset_id, filter: ["countryiso3code:eq:DEU", "date:gte:2010", "date:lte:2020"], sort: "date:asc")- The same call over REST:
GET /api/v1/public/datasets/6217e2ba-f66a-45c9-8d0d-953daeb0670d/data?filter=countryiso3code:eq:DEU&filter=date:gte:2010&filter=date:lte:2020&sort=date:asc
- Answer
-
| Year | Value (%) |
|---|
| 2010 | 7.043 |
| 2011 | 5.967 |
| 2012 | 5.372 |
| 2013 | 5.316 |
| 2014 | 4.979 |
| 2015 | 4.612 |
| 2016 | 4.104 |
| 2017 | 3.781 |
| 2018 | 3.384 |
| 2019 | 3.163 |
| 2020 | 3.881 |
All 11 rows, as they are returned.
11 rows, about 356 tokens in TOON.
Straight from the source API
What you have to know before you can write it: the indicator code SL.UEM.TOTL.ZS, the entity code DEU and the provider's own date syntax date=2010:2020.
GET https://api.worldbank.org/v2/country/DEU/indicator/SL.UEM.TOTL.ZS?date=2010:2020&format=json&per_page=100
The answer arrives as nested JSON that repeats the indicator name and the country on every row:
{"indicator":{"id":"SL.UEM.TOTL.ZS","value":"Unemployment, total (% of total labor force) (modeled ILO estimate)"},"country":{"id":"DE","value":"Germany"},"countryiso3code":"DEU","date":"2020","value":3.881,"unit":"","obs_status":"","decimal":1}
11 rows, 2,798 bytes, about 700 tokens in JSON.
Same 11 rows, about 2 times fewer tokens. The bigger difference is that one side takes a question and the other takes three codes.
If you paste the whole table instead, the same period for all 265 entities is 2,915 rows and about 181,731 tokens. That is the 510 times figure, and it answers a different question.
Measured on 2026-09-04 against Unemployment Rate by Country (World Bank), published by World Bank. The source API call was measured on 2026-09-09. Token counts are approximations at about four characters per token.
Source dataset: Unemployment Rate by Country (World Bank)
What the data layer guarantees
- Normalized at ingestion
- Cleaned, schema-detected, entity-resolved and shadow-columned when it arrives, not when you query it. That is what makes a join across two publishers a filter instead of a project.
- Every number cites its source
- Each dataset links to the primary source it came from and each row keeps its publisher and its period. Nothing is generated, and a number that cannot be traced does not ship.
- Always up to date
- A self-healing engine refreshes on a schedule or on demand, and the builder picks which. When a dataset changes shape, the change is written to a changelog your app can read.
- Open and model-agnostic
- Public data is free to browse and query without an account, and no provider is wired in. Connect Claude, ChatGPT, Gemini or anything else that speaks MCP.
Start with a question
Open an app, read the data, or connect your AI through MCP. Looking costs nothing and needs no account.