What Is Querio? Git-Native Analytics for Data Teams

Querio is git-native analytics for data teams and their AI agents. Your metrics, context, saved queries, dashboards and permissions live as plain files in a git repo you own. Agents can do the work: they propose changes, a person approves them in a pull request, and every AI tool in the company answers from the same definitions.

Who makes Querio?

Querio Ltd makes Querio. Rami Abi Habib (CEO) and I, Javier Bonilla (CTO), founded the company. We relaunched on October 27, 2026, with $5 million from BMW i Ventures, LAUNCH and other investors.

People now ask data questions in Claude, in Slack and wherever else they work, and they get three different revenue numbers. That's the problem we built Querio to fix.

What does Querio mean?

The name reads like "query", which is most of what people do with it. It also comes from "quiero", Spanish for "I want", a nod to my roots in Mexico. People want to query their data, and Querio lets them.

Is Querio the same as the Querio Chrome extension?

No. Search "querio" and you'll also see a Chrome extension called "Querio - GraphQL and HTTP requests inspector". It's someone else's developer tool for inspecting network requests, and it has nothing to do with us.

What does git-native analytics mean?

It means the repo is the system of record, not our database. Here's what lives there as plain files:

• context (your catalog and knowledge base)
• metric definitions
• saved queries
• dashboards
• automations
• notebooks
• permission policies

Sync runs both ways. Edit something in the Querio app and it's written back to the repo. Edit a file in VS Code, Cursor, Claude Code or CI and it syncs into Querio. If you ever leave, you keep the repo.

If you use dbt, Querio reads dbt's semantic_manifest.json and uses your MetricFlow metrics as they are, with no second copy to keep in step. You don't need dbt, though. Without it, you define metrics in Querio and they're committed to the repo as MetricFlow-format files. More on that in our guide to the dbt Semantic Layer.

How does a change reach the business?

Through your own review process. Here's what happens when someone renames a dbt model, say fct_orders to fct_sales. CI flags every dashboard, saved query and context file that still points at the old name, and the pull request can't pass until they're fixed. Nobody finds out from a broken chart on Monday morning.

Agent changes work the same way:

1. An agent proposes an update to your context.
2. A person saves it.
3. Saving context or promoting a dashboard opens a pull request on your repo, under your branch protections and reviewers.
4. Merging it makes the change live in Querio.

Dashboards carry a level: experimental, team or trusted. Whoever can merge to main can promote one to trusted.

Where can people and agents use Querio?

Wherever they already work:

• the Querio web app
• Slack, through the Querio agent
• Claude, ChatGPT and custom agents, through our MCP server
• Claude Code, Cursor and CI, through our CLI
• the API
• inside your own product, as embedded analytics

Through the MCP server, an agent can get a metric by name, run SQL and propose a context change as a pull request, among other things. It runs as you and sees only what you can see. On Snowflake, BigQuery and Databricks, each person connects with their own OAuth login, so the warehouse's own access rules apply. If you're setting this up for Claude, start with how to connect Claude to your data warehouse with MCP.

Every question runs as a reactive SQL and Python notebook in a sandbox. Analysts can open the code behind an answer, and an agent can keep a dataframe and build on its last result.

Who is Querio for?

Data leads and analytics engineers who own the warehouse, keep the dbt models, and field the business's questions. Querio connects to the warehouse or database your data already lives in, such as Snowflake, BigQuery, Databricks, Redshift, Postgres or ClickHouse.

Coming from Tableau, Looker or Power BI? You don't start from scratch: unshackle, our free open-source migration tool, moves your metrics, dashboards and permissions into your repo. Weighing Querio against a notebook tool? Read Querio vs Hex. Against other semantic layers? Here are the top semantic layer tools.

How much does Querio cost?

There's no per-seat pricing. Every plan has unlimited users and data connections. Only AI model calls are metered, as credits. One credit is one call.

• Free: $0, free forever. 500 AI credits a month, with the MCP server and CLI included.
• Core: $2,499 a month, with 10,000 AI credits a month.
• Enterprise: custom pricing.

The full breakdown is on our pricing page.

Is Querio secure?

Querio is SOC 2 Type II compliant, and you can request the report from our Trust Center. We're HIPAA compliant and sign BAAs, and we're GDPR and ITAR compliant.