> For the complete documentation index, see [llms.txt](https://docs.algenta.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.algenta.ai/sdks/algenta-cli.md).

# algenta CLI (local)

`algenta` is the **local-first** command. It runs entirely on your machine — a deterministic runtime demo, CSV import previews, instant schema mapping, and local queries — with no server and no API key. It is the zero-friction way to feel Algenta before you stand up the engine. For the full control-plane surface (keys, connectors, agent runs, repositories, deployments), use the [`de` CLI](/sdks/cli.md), which talks to the `/v1` HTTP API.

{% hint style="info" %}
The `algenta` command ships today with `algenta-core`. Install it with `pip install algenta-core` — **not** `pip install algenta`, which is not published yet. The account and agent commands (`login`, `whoami`, `runtime doctor`, `mcp`) are rolling out; the local commands below work now.
{% endhint %}

## algenta vs de

|                  | `algenta`                                               | `de`                                                               |
| ---------------- | ------------------------------------------------------- | ------------------------------------------------------------------ |
| Install          | `pip install algenta-core`                              | `pip install algenta-sdk`                                          |
| Where it runs    | locally, in-process                                     | against the `/v1` API (self-hosted or hosted)                      |
| Needs an API key | no (local commands); `algenta login` for account work   | yes                                                                |
| Best for         | offline previews, the runtime demo, quick local queries | the full platform: keys, data, agents, repositories, control plane |

## Install and check

```bash
pip install algenta-core
algenta version
```

`algenta version` prints the installed version and the primary runtime path (`Runtime.resolve()` → `Runtime.query()`), plus the secondary local modes.

## Local commands

These run on your machine and need no key or network.

| Command                                            | What it does                                                          |
| -------------------------------------------------- | --------------------------------------------------------------------- |
| `algenta demo`                                     | Interactive local runtime demo (register → resolve → verify → query). |
| `algenta demo --list`                              | List the selectable demo scenarios and exit.                          |
| `algenta demo --scenario <id>`                     | Run one named scenario non-interactively.                             |
| `algenta demo --all --no-interactive`              | Run every scenario in sequence, no prompts.                           |
| `algenta import <file.csv> [--name <n>]`           | Branded local import preview for one source.                          |
| `algenta import a.csv b.csv`                       | Bundle preview across several sources (finds relationships).          |
| `algenta map <file1.csv> [file2.csv ...]`          | Instant local schema mapping and join-key discovery.                  |
| `algenta query "<text>" <file.csv> [more.csv ...]` | Deterministic local query helper.                                     |
| `algenta simulate`                                 | Run a quick local Monte Carlo simulation.                             |
| `algenta benchmark`                                | Show local speedup versus a raw-Python baseline.                      |
| `algenta upgrade`                                  | Print how to switch to direct API mode.                               |
| `algenta version`                                  | Show version and capabilities.                                        |

### `algenta demo`

The built-in scenario keys are `product_line`, `region`, `canada_filter`, and `guardrail`.

{% code title="algenta-demo.sh" %}

```bash
algenta demo                                  # interactive picker (in a TTY)
algenta demo --list                           # list scenarios and exit
algenta demo --scenario region --no-interactive
algenta demo --all --no-interactive           # run them all
```

{% endcode %}

### `algenta import`

`--name` is only valid when importing a single source. Import ends by printing a copy-pasteable `Runtime()` snippet.

{% code title="algenta-import.sh" %}

```bash
algenta import orders.csv --name orders
algenta import orders.csv financial.csv ratings.csv
```

{% endcode %}

### `algenta map`

{% code title="algenta-map.sh" %}

```bash
algenta map sales.csv orders.csv
```

{% endcode %}

Add a second source to surface cross-table join keys and confidence scores.

### `algenta query`

The query text comes first, then one or more files.

{% code title="algenta-query.sh" %}

```bash
algenta query "total revenue by customer" sales.csv
algenta query "top 10 orders by amount" orders.csv
```

{% endcode %}

### `algenta simulate` and `algenta benchmark`

{% code title="algenta-simulate.sh" %}

```bash
algenta simulate     # quick local simulation (revenue=100,000 cost=70,000 runs=10,000)
algenta benchmark    # local speedup vs a Python baseline
```

{% endcode %}

### `algenta upgrade`

Prints the exact steps to move from the in-process runtime to direct API mode — the base URL to set, where to get a key, and the before/after `Runtime(mode="api", ...)` code.

{% code title="algenta-upgrade.sh" %}

```bash
algenta upgrade
```

{% endcode %}

## Account and agent commands (rolling out)

These connect the local CLI to your account and to agent tooling. They are shipping incrementally; until they land on your install, use the [`de` CLI](/sdks/cli.md) and [API-key authentication](/getting-started/authentication.md).

| Command                      | What it does                                                                                           |
| ---------------------------- | ------------------------------------------------------------------------------------------------------ |
| `algenta login`              | Browser device login — approve a code at `accounts.thyn.ai`, then the CLI stores a machine-scoped key. |
| `algenta login --no-browser` | Print the verification URL and code instead of opening a browser (headless / SSH).                     |
| `algenta whoami`             | Confirm you are signed in (prints the stored key's prefix).                                            |
| `algenta runtime doctor`     | Diagnose the bundled in-process native runtime.                                                        |
| `algenta mcp`                | Start the Algenta MCP server for local AI clients.                                                     |

{% code title="algenta-login.sh" %}

```bash
algenta login
algenta login --no-browser
algenta whoami
```

{% endcode %}

See [Log in from the terminal](/getting-started/login.md) for the full device-login walkthrough and [Install from the MCP Registry](/sdks/mcp-registry.md) for connecting the MCP server to Cursor, Claude, and Codex.

## Related

{% content-ref url="/pages/UfXsQe4akQMvhWvf2CNv" %}
[CLI reference](/sdks/cli.md)
{% endcontent-ref %}

{% content-ref url="/pages/3Np7OG43tmtl7yxBj9jn" %}
[Log in from the terminal](/getting-started/login.md)
{% endcontent-ref %}

{% content-ref url="/pages/paee6eT9BsPydUAeJEsQ" %}
[Install from the MCP Registry](/sdks/mcp-registry.md)
{% endcontent-ref %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.algenta.ai/sdks/algenta-cli.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.