# Alchemist

> Alchemist is a SAS migration toolchain by T1A. It inventories a legacy SAS estate,
> maps the dependencies between its jobs and datasets, and converts SAS code to
> PySpark for Databricks and Snowflake. It is driven either through its own
> interface or from a coding agent over MCP.

The product has three surfaces. The **Analyzer** builds an inventory of an existing
SAS environment and a visual dependency graph over it, so a migration can be
sequenced before any code is rewritten. The **Converter** translates SAS into
PySpark using tree pattern matching and a templating engine, resolving SAS metadata
(libraries, formats, macro variables) as it goes. The **agent layer** exposes that
same conversion to coding agents as an MCP server and as Skills.

Alchemist runs either as a hosted service at `api.getalchemist.io`, or fully
air-gapped inside a customer network as a Docker image, where no code leaves the
customer's environment.

## Product

- [Alchemist](https://getalchemist.io/): The Analyzer (inventory and dependency mapping of a SAS estate), the Converter (SAS to PySpark), and the migration methodology behind them.
- [Agents — MCP and Skills](https://getalchemist.io/agents/): Driving Alchemist conversions from a coding agent. Works with Databricks Genie, Snowflake Cortex, Claude Code, or any agent that speaks MCP.

## Documentation

Full documentation, 33 pages: https://docs.getalchemist.io/

- [Quick start](https://docs.getalchemist.io/quick_start.html): Getting a first analysis and conversion running.
- [Configuration](https://docs.getalchemist.io/configuration/index.html): Project, analysis, and notification configuration.
- [Conversion configuration](https://docs.getalchemist.io/configuration/conversion/index.html): Node filters, templates, and generative-AI-assisted conversion.
- [SAS dialect](https://docs.getalchemist.io/configuration/dialects/sas/index.html): The source dialect and its parser.
- [Spark dialect](https://docs.getalchemist.io/configuration/dialects/spark/index.html): PySpark as a conversion target, and its templates.
- [Databricks dialect](https://docs.getalchemist.io/configuration/dialects/dbr/index.html): Databricks as a conversion target.
- [Prophecy dialect](https://docs.getalchemist.io/configuration/dialects/prophecy/index.html): Prophecy as a conversion target.
- [Analyzing](https://docs.getalchemist.io/usage/analyzing.html): Running an analysis over a SAS codebase.
- [Converting](https://docs.getalchemist.io/usage/converting/index.html): Running conversions, and the documented nuances of SAS-to-Spark and SAS-to-Prophecy output.
- [AST](https://docs.getalchemist.io/usage/ast.html): The abstract syntax tree Alchemist builds from SAS source.
- [Preparing a folder structure](https://docs.getalchemist.io/usage/preparing_folder_structure.html): How to lay out a codebase before analysis.
- [Licensing](https://docs.getalchemist.io/licensing/index.html): Licensing model and the metrics it is measured on.
- [Changelog](https://docs.getalchemist.io/changelog.html): Release history.

## Deployment

- **Alchemist Cloud** — the agent connects to `api.getalchemist.io` over the public internet. No infrastructure to run.
- **Air-gapped (Docker)** — the full stack is deployed inside the customer's own network (`docker load -i alchemist.tar`). No data leaves the environment.

## Pricing

Pricing is not published on the site. Licensing is documented at
https://docs.getalchemist.io/licensing/index.html and the metrics it is measured on
at https://docs.getalchemist.io/licensing/metrics.html. For a quote, contact the team
through https://getalchemist.io/#contact.

## Machine-readable resources

- `/llms.txt` — this file
- `/mcp` — agent-oriented index of the product. A static document, not an MCP protocol endpoint; it states where the real MCP server runs.

## Links

- Website: https://getalchemist.io/
- Documentation: https://docs.getalchemist.io/
- Live demo: https://live.getalchemist.io/
- Application: https://app.getalchemist.io/
- MCP / API endpoint: https://api.getalchemist.io/
- Company: https://t1a.com/
- LinkedIn: https://www.linkedin.com/company/tier-one-analytics/
- Contact: https://getalchemist.io/#contact