Engine
One engine
Every action is a command with an id. The desktop app, the command line and the MCP server all call the same commands, so a step you take in one place works the same way in the others.
- query.run
- template.render
- dashboard.publish
In private development
None of these are released yet. The desktop app, the command line and the MCP server are in development; the GitHub App is planned.
Integrations
Where you use Emolysis.
Emolysis is one engine with one set of commands. You reach it from a desktop app, from the emo command line, from any editor or coding agent that supports MCP, and from checks on your GitHub pull requests (planned). A command that works in one place works the same way in all of them.
The surfaces
01
In development
Desktop app
The full engine with a workspace view, for macOS, Windows and Linux.
- Runs on
- macOS, Windows, Linux
- Includes
- Engine, DuckDB, emo
- Builds
- Signed (planned)
02
In development
emo CLI
Every Emolysis command from the terminal, for scripts and CI.
- Commands
- init, connect, check
- Output
- JSON for scripts
- Same as
- App and MCP
03
In development
MCP server
Emolysis inside any editor or coding agent that supports MCP.
- Transport
- Standard input and output
- Account
- Needed, Free plan included
- Tools
- One per command, except machine tokens
04
Planned
GitHub App
Checks on every pull request: lint, golden tests, cost estimate and a preview link.
- Checks
- Lint, tests, cost, preview
- Reads
- Contents, metadata
- Also as
- A CI workflow
Same commands
The same step, in three places.
Every step is a command. This is how four of them look on each surface. Pull request checks, which are planned, run in GitHub, on the pull request itself.
| Step | Desktop app | emo command line | MCP server |
|---|---|---|---|
| Connect a platform | Connect, then Databricks | emo connect databricks | The platform_connect tool |
| Ask a business question | Type the question | emo run dashboard.draft --params '<json>' | The dashboard_draft tool |
| Run any command | Ask the agent | emo run <id> --params '<json>' | The command's tool, such as template_render |
| Deploy a green pipeline | Deploy | emo run pipeline.deploy --params '<json>' | The pipeline_deploy tool |
Read next
Questions
Integrations, answered.
Where it runs, which systems, and how GitHub fits. More in the FAQ.
Where can I use Emolysis?
From the desktop app, the emo command line, any editor or coding agent that supports MCP, and checks on your GitHub pull requests. They share one engine and the same commands. The desktop app, the command line and the MCP server are in development, and the GitHub App is planned. None is released yet.
Can I use Emolysis from my editor or coding agent?
Not yet: nothing is released. Emolysis includes an MCP server, emo mcp, so an editor or coding agent that supports the Model Context Protocol can use the same commands as the desktop app, with one configuration entry.
Which operating systems does Emolysis run on?
macOS, Windows and Linux. The desktop app is built for each: release builds for macOS and Windows are signed, and Linux gets an unsigned AppImage, checked against its published checksum. The emo command line and the MCP server run on the same three systems. There is nothing to download yet.
How does Emolysis work with GitHub?
Your pipelines live in your own repository as a plain project. A planned GitHub App checks each pull request with lint, golden tests and a cost estimate, and links a preview dashboard. Green checks become your deploy gate. You can use Emolysis without GitHub too.
Do all the surfaces behave the same way?
Yes. Every action is a command, available the same way in the app, the command line and the MCP server, with the same lint rules, templates and golden tests; only the machine-token commands stay off the MCP server. Every surface needs an Emolysis account, local files included, and the same sign-in serves them all.
Nothing is released yet
Start where you work.
Early access is by invitation, starting with data engineering teams on Databricks.