Skip to main content

Documentation index: llms.txt. This page is also available as markdown: append .md to this URL or send Accept: text/markdown.

Coalesce Documentation

Coalesce is the data operating layer where humans and AI agents build, govern, and observe data together. It brings transformation, cataloging, and monitoring into one governed platform, so every person and every agent builds to the same standards, with the same context.

The Coalesce Platform​

ProductWhat it doesStart here
Coalesce TransformBuild, automate, and govern data transformations with agents, in the web app, or directly through code.What is Transform?
Coalesce CatalogDiscover, document, and govern every data asset with AI.What is Catalog?
Coalesce QualityMonitor, test, and validate data with observability built into the pipeline.What is Quality?

All three share one metadata layer, so lineage, ownership, and test results follow your data from the first draft to production. See How Coalesce Fits Together.

How You Build​

Your coding agent writes your pipeline as SQL files in Git. You review every change in Coalesce Desktop, and Coalesce deploys it through the same governed process as the rest of your work.

You describe, your agent builds

Describe and build

Tell your agent what you need. It writes SQL Nodes as files in Git and runs them in your development schema.

Set up your agent →
You review

Review in Coalesce Desktop

Watch Nodes appear as your agent works. Check what changed, the SQL, and lineage before anything ships. Not right? Ask for changes and review again.

Review agent work →
Coalesce ships and observes

Ship and observe

Deploy through governed plans and schedule Jobs. Coalesce Catalog documents what shipped, and Coalesce Quality monitors it.

How it fits together →
One shared metadata layerCoalesce TransformCoalesce CatalogCoalesce Quality
  1. Describe and build. Tell your agent what you need. It writes SQL Nodes as files in Git and runs them in your development schema. See Set Up Your Agent.
  2. Review in Coalesce Desktop. Watch Nodes appear as your agent works. Check what changed, the SQL, and lineage before anything ships. Not right? Ask for changes and review again. See Review Agent Work.
  3. Ship and observe. Deploy through governed plans and schedule Jobs. Coalesce Catalog documents what shipped, and Coalesce Quality monitors it. Coalesce Transform, Catalog, and Quality share one metadata layer. See How Coalesce Fits Together.

Every step works without an agent too. Write the SQL yourself, or build the same pipeline in the web app.

Start Building​

Already use a coding agent? Paste this into Claude Code, Cursor, Codex, or any other. It installs the Coalesce platform skills, gives you the command to connect Transform MCP, checks your Workspace, and reports back. It doesn't change anything in your warehouse or your Environments.

Read https://docs.coalesce.io/docs/get-started/set-up-your-agent.md and follow the instructions to set yourself up for Coalesce.

Coalesce Apps​

Add your team's own tools to Coalesce Desktop. Apps live in your Workspace's Git repository and open as tabs in Desktop, so every developer gets the same tools. Install one from the App Marketplace as it rolls out, such as Build with Agents, which turns a spec into a governed pipeline, or have your agent build one. See Coalesce Apps.

Get Help​

For your agent: add .md to any page's URL for a Markdown copy, or point your agent at llms.txt, which indexes every page by product.