> ## Documentation Index
> Fetch the complete documentation index at: https://docs.lilfella.app/llms.txt
> Use this file to discover all available pages before exploring further.

# Using Fella

> The complete path from a folder on disk to an auditable answer.

Fella is a local-first desktop app for asking plain-language questions about files in a folder. It keeps model-written prose separate from the tool results and checks that support it.

## Before you start

* Open a folder containing a [supported table or document format](/developer-platform/using-fella/files).
* [Connect a model provider](/developer-platform/using-fella/models) with your own API key.

After this workflow, you can inspect the sources, computations, and verification findings behind an answer.

## The four-part workflow

<Steps>
  <Step title="Open a workspace">
    Choose a folder, drag it into the app, or use `/open`. Fella catalogs supported files and prepares tables and documents for read-only tools.
  </Step>

  <Step title="Ask a question">
    Ask about tables, documents, trends, comparisons, joins, or correlations. SQL is the default computation path; Python is the escape hatch.
  </Step>

  <Step title="Inspect the evidence">
    Open **Evidence** to see tool steps, queries or arguments, results, timings, and verification findings. Charts render below the answer.
  </Step>

  <Step title="Refine the workspace">
    Ask a follow-up, add `fella.md` context, correct vocabulary, switch models, or open another conversation.
  </Step>
</Steps>

## Choose a guide

<CardGroup cols={2}>
  <Card title="Open a folder" icon="folder-open" href="/developer-platform/using-fella/open-folder">
    Workspace boundaries, scanning, `.fellaignore`, and the catalog.
  </Card>

  <Card title="Ask questions" icon="message-circle" href="/developer-platform/using-fella/asking-questions">
    How to phrase questions and how Fella chooses SQL, documents, Python, and charts.
  </Card>

  <Card title="Read the evidence" icon="search-check" href="/developer-platform/using-fella/evidence">
    Follow every computation and understand the verification checklist.
  </Card>

  <Card title="Understand files" icon="files" href="/developer-platform/using-fella/files">
    Supported formats, table names, documents, and rescanning.
  </Card>

  <Card title="Add context and memory" icon="brain" href="/developer-platform/using-fella/context-and-memory">
    Teach Fella your vocabulary without turning context into fake evidence.
  </Card>

  <Card title="Choose a model" icon="cpu" href="/developer-platform/using-fella/models">
    Hosted providers, custom endpoints, and credential storage.
  </Card>

  <Card title="Manage sessions" icon="panels-top-left" href="/developer-platform/using-fella/sessions">
    Tabs, the history sidebar, focus mode, retrying, and updates.
  </Card>

  <Card title="Commands and shortcuts" icon="book-open" href="/developer-platform/using-fella/commands">
    The command palette, slash commands, and keyboard reference.
  </Card>
</CardGroup>

## Product boundary

The base agent has no tool that writes, moves, or deletes workspace files. The model chooses read-only tools; the data engine runs their computations and records evidence. Verification checks the completed answer, but it does not prove that a valid query interpreted an ambiguous question correctly.

<Warning>
  Python runs inside an embedded WASM capability boundary. Hosted providers
  receive prompts and tool results for the question. Review [privacy and
  security](/developer-platform/using-fella/privacy) before using sensitive data.
</Warning>

<Note>
  Fella answers questions about a workspace. It is not a general task agent, file manager, coding assistant, or cloud service.
</Note>

<CardGroup cols={2}>
  <Card title="The harness, engine, and model" icon="layers" href="/concepts/triad">
    The ownership boundaries behind the user workflow.
  </Card>

  <Card title="Privacy and security" icon="shield-check" href="/developer-platform/using-fella/privacy">
    Provider egress, credentials, and the Python boundary.
  </Card>
</CardGroup>