# fella

> Documentation for Fella, a local-first desktop app for asking auditable questions about your own files.

- [Fella](https://docs.lilfella.app/index.md): Ask questions about your own files and inspect the work behind tool-backed answers.
- [Getting started](https://docs.lilfella.app/getting-started.md): Install Fella, connect a model, open a folder, and inspect a tool-backed answer.
- [Examples](https://docs.lilfella.app/examples.md): Task recipes for inspecting, comparing, joining, searching, and charting personal data.
- [Troubleshooting](https://docs.lilfella.app/troubleshooting.md): Diagnose installation, workspace, model, query, and Python problems.
- [Limitations](https://docs.lilfella.app/limitations.md): Current boundaries for correctness, formats, scale, security, and platforms.
- [Using Fella](https://docs.lilfella.app/developer-platform/using-fella.md): The complete path from a folder on disk to an auditable answer.
- [Open a folder](https://docs.lilfella.app/developer-platform/using-fella/open-folder.md): Choose the workspace Fella can inspect and understand how the catalog is built.
- [Files and documents](https://docs.lilfella.app/developer-platform/using-fella/files.md): Supported formats, table naming, document search, and rescanning.
- [Context and memory](https://docs.lilfella.app/developer-platform/using-fella/context-and-memory.md): Teach Fella your vocabulary and understand what it remembers about a workspace.
- [Ask questions](https://docs.lilfella.app/developer-platform/using-fella/asking-questions.md): How to phrase questions and how Fella chooses deterministic tools.
- [Read the evidence](https://docs.lilfella.app/developer-platform/using-fella/evidence.md): Inspect the tools, queries, rows, timings, and verification checks behind an answer.
- [Models and providers](https://docs.lilfella.app/developer-platform/using-fella/models.md): Connect a hosted provider or custom OpenAI-compatible endpoint with your own key.
- [Tabs, history, and focus](https://docs.lilfella.app/developer-platform/using-fella/sessions.md): Keep conversations separate, reopen past work, and use the compact focus view.
- [Experimental MCP](https://docs.lilfella.app/extensions-and-mcp.md): The extension boundary in Fella's lean personal release.
- [The harness, engine, and model](https://docs.lilfella.app/concepts/triad.md): The three-part architecture behind Fella's model loop, local analytics, and evidence.
- [The harness](https://docs.lilfella.app/concepts/harness.md): How Fella controls model context, tool calls, evidence, verification, and stopping.
- [The analytics engine](https://docs.lilfella.app/concepts/engine.md): How Fella catalogs files and produces local SQL, document, Python, and chart results.
- [Built-in tools](https://docs.lilfella.app/concepts/tools.md): The seven built-in tools, their purposes, and their actual guardrails.
- [The model layer](https://docs.lilfella.app/concepts/model-layer.md): Current model providers, protocols, credentials, selection, and network boundaries.
- [Verification and evidence](https://docs.lilfella.app/concepts/verification.md): How deterministic answer checks and optional model second opinions differ.
- [Privacy and security](https://docs.lilfella.app/developer-platform/using-fella/privacy.md): Understand the read-only boundary, network behavior, credentials, and Python guard rails.
- [Commands and shortcuts](https://docs.lilfella.app/developer-platform/using-fella/commands.md): Complete reference for Fella's slash commands and keyboard controls.
- [Benchmarks](https://docs.lilfella.app/other-resources/benchmarks.md): Reproduce and interpret Fella's dated harness-comparison measurements.
- [Evaluation and testing](https://docs.lilfella.app/other-resources/evaluation.md): Choose the test layer that matches the claim you are changing.
- [Contributing](https://docs.lilfella.app/other-resources/contributing.md): Make a focused app change and verify the claims it affects.