Status
In private development. Early access by invitation. The names, options and formats on this page can change before the first release.
emo is the command line for the Emolysis engine. It is built for terminals, scripts and CI, and it runs the same commands as the desktop app and the MCP server.
Commands
| Command | What it does |
|---|---|
emo init <name> | Create a pipeline project: models/, pipelines/, tests/, emolysis.yml |
emo connect <platform> | Connect a platform, for example emo connect databricks |
emo login, emo logout, emo whoami | Sign in, sign out, show who is signed in |
emo check | Lint, golden tests and a cost estimate for the project, for CI |
emo token create, list, revoke | Manage your organization's machine tokens for CI |
emo dashboard refresh | Refresh a dashboard's data; --publish publishes it |
emo mcp | Start the MCP server over standard input and output |
emo doctor | Check your setup; exits with an error when a required check fails |
emo version | Print the engine version |
emo commands | List every command id, with --schemas for their parameters |
emo run <id> --params '<json>' | Run any command by its id with JSON parameters |
Deploying is emo run pipeline.deploy.
Run emo login to sign in. Every command except emo version, emo doctor, emo commands and the sign-in commands needs an Emolysis account, local files included. A command you run while signed out starts the sign-in flow: emo prints a short code, and you confirm it in the browser, where you sign in with your GitHub, Microsoft or Google account. The code expires after 15 minutes.
A session
Shell
emo init orders
emo connect databricks
emo checkEach line maps to a step in How Emolysis works. The Quickstart walks through the same session in the app.
Commands by id
Every action in Emolysis is a command with an id of the form <area>.<verb>, such as query.run, dialect.transpile, dialect.lint, template.list, template.render, template.test or dashboard.publish. emo commands lists them all. emo run calls any of them:
Shell
emo run dialect.transpile --params '{"sql": "select 1", "write": "databricks"}'emo run prints the result as JSON and exits with status 0, or prints an error object and exits with status 1. The error object carries one of nine error codes; see Error codes. That makes emo run the dependable way to use Emolysis from scripts and CI.
The short commands, such as emo version and emo doctor, are shortcuts to registered commands and add no logic of their own. A command available on one surface is available on all of them: the app, the CLI and the MCP server.
Exit codes
| Result | Exit status |
|---|---|
| A command returned a result | 0 |
| A command returned an error object | 1 |
emo doctor, a required check fails | 1 |
In scripts, use the status and read the error's code, not its message, to decide what to do. The messages are for people.
Skills in the app and in coding agents
The same engine offers a few higher-level requests as skills, with slot schemas so the agent asks only for what is missing:
| Skill | What it does |
|---|---|
/dashboard "<question>" | Find sources, propose a semantic model, build the gold table, chart it, publish on request |
/pipeline <shape> from <source> to <target> | Fill the template for a shape, run its tests, deploy on request |
/connect | Connect a platform |
/migrate <table> to <platform> (planned) | Translate a pipeline to another platform and re-run its tests |
/explain <pipeline> | Describe a pipeline's lineage and cost in plain language |
CI
In CI, use emo run for anything scripted and the GitHub checks for pull requests. Because the commands are the same everywhere, the steps you run on your machine are the steps a runner runs.
Installing
emo comes with the desktop app. A package on PyPI is planned; see Installation. The name emolysis is reserved on PyPI with an empty placeholder.
Status
The command names, the options and the output formats can change before the first release. The command id form, the two exit statuses of emo run and the nine error codes are fixed.