libre-webui 0.16.1 → 0.18.0

package/README.md

@@ -25,7 +25,7 @@ Self-hosted. Provider-flexible. Apache 2.0.
   <a href="https://librewebui.org"><strong>Website</strong></a> ·
   <a href="https://docs.librewebui.org"><strong>Documentation</strong></a> ·
   <a href="https://github.com/libre-webui/libre-webui/releases"><strong>Download</strong></a> ·
-  <a href="#start-in-one-command"><strong>Quick start</strong></a>
+  <a href="#quick-start"><strong>Quick start</strong></a>
 </p>
 
 <br>
@@ -36,148 +36,203 @@ Self-hosted. Provider-flexible. Apache 2.0.
 
 ---
 
-## Freedom should survive success
-
-An AI interface can be self-hosted today and still become a gatekeeper tomorrow. Libre WebUI is built so your freedom does not depend on a company remaining benevolent.
-
-Run it. Inspect it. Change it. Fork it. Connect it to something else. Or leave it behind. The code is Apache 2.0, local inference is the default path, remote providers are opt-in, and the project charter puts independence and inclusive participation into writing.
+## What Libre WebUI is
 
-> **Open source is a license. Libre is an operating principle.**
+A complete AI workspace you run yourself. Chat with local models through
+[Ollama](https://ollama.com) or with any provider you choose to connect, give a
+model a sandboxed workspace where it can actually build something, search your
+own documents, generate images and speech, and keep every conversation on
+hardware you control.
 
-Do not trust the pitch. Inspect the [license](https://github.com/libre-webui/libre-webui/blob/main/LICENSE), [charter](https://github.com/libre-webui/libre-webui/blob/main/CHARTER.md), [source](https://github.com/libre-webui/libre-webui), and [CI](https://github.com/libre-webui/libre-webui/actions).
+The interface works like a browser. **Home** is your launcher; chats, Work
+sessions, and pages open as tabs beside it; `⌘K` searches everything you have.
+Nothing is buried three menus deep.
 
-| What stays yours      | Libre WebUI's approach                                                              | Verify it                                                                  |
-| --------------------- | ----------------------------------------------------------------------------------- | -------------------------------------------------------------------------- |
-| **Data path**         | Self-hosted by default; remote inference only when you select it                    | [Deployment docs](https://docs.librewebui.org/DOCKER)                      |
-| **Model choice**      | Local Ollama plus optional provider plugins and OpenAI-compatible endpoints         | [Plugin architecture](https://docs.librewebui.org/PLUGIN_ARCHITECTURE)     |
-| **Code**              | Use, modify, redistribute, and fork under Apache 2.0                                | [License](https://github.com/libre-webui/libre-webui/blob/main/LICENSE)    |
-| **Project direction** | The charter rejects funding control over the roadmap, license, or community         | [Charter](https://github.com/libre-webui/libre-webui/blob/main/CHARTER.md) |
-| **Participation**     | Contributors and users are welcome regardless of background, identity, or geography | [Charter](https://github.com/libre-webui/libre-webui/blob/main/CHARTER.md) |
-| **Exit**              | Your installation and your fork do not need our permission                          | [Source](https://github.com/libre-webui/libre-webui)                       |
+It is Apache 2.0, ships without telemetry, and is built so you can leave. If
+this project ever stops serving you, your installation and your fork do not need
+anyone's permission.
 
-## Start in one command
+## Quick start
 
-With [Node.js 22.22 or newer](https://nodejs.org):
+You need [Node.js 22.22 or newer](https://nodejs.org). Docker is optional — only
+[Work](#work-one-persistent-environment-per-task) requires it.
 
 ```bash
 npx libre-webui
 ```
 
-Open [http://localhost:8080](http://localhost:8080). The first account created on a fresh install becomes the administrator.
+Open [http://localhost:8080](http://localhost:8080). The first account created
+on a fresh install becomes the administrator.
 
-For private local inference, install [Ollama](https://ollama.com) and pull a model:
+For private local inference, install [Ollama](https://ollama.com) and pull a
+model:
 
 ```bash
-ollama pull gemma3:4b
+ollama pull gemma4:12b
 ```
 
-That is enough to start. Cloud accounts are not required. When you do want a remote model, add only the provider you choose.
-
-Chat works without Docker. The new **Work** mode additionally needs Docker on
-the machine running the Libre WebUI backend. If Docker is missing, Libre WebUI
-still starts normally and Work shows a clear **Runtime unavailable** state
-instead of running commands on the host.
+That is enough to start. No cloud account is required. When you do want a remote
+model, add only the provider you choose.
 
-## One workspace. Your choices.
+If Docker is missing, Libre WebUI starts normally and Work reports a clear
+**Runtime unavailable** state rather than running commands on your host.
 
-Libre WebUI turns a model endpoint into a complete working environment without taking ownership of the stack around it.
+## What you can do
 
-|                                 | What you can do                                                                                                                              |
-| ------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------- |
-| **Talk to the model you want**  | Stream conversations through local Ollama, Ollama Cloud, bundled providers, or compatible endpoints. Discover and manage models from the UI. |
-| **Give every task a workspace** | Let a tool-capable model inspect files, edit code, run commands, and preview an app in a persistent, task-scoped Docker workspace.           |
-| **Bring your own context**      | Chat with PDF and plain-text documents using keyword retrieval or optional semantic embeddings.                                              |
-| **Turn answers into artifacts** | Preview sandboxed HTML, SVG, JSON, code, and multi-file artifacts beside a normal chat.                                                      |
-| **Create with more than text**  | Generate images, use provider-backed speech, and build reusable personas with prompts, parameters, and memory.                               |
-| **Make the interface yours**    | Keep light or dark mode across refreshes, apply an adaptive accent theme, and work in 25 languages/locales including Arabic RTL.             |
-| **Operate it your way**         | Use local accounts, roles, optional OAuth, Docker, Kubernetes, npm, or the Electron desktop client.                                          |
-| **Extend without lock-in**      | Configure chat, embeddings, image generation, and text-to-speech providers through the plugin layer.                                         |
+|                                    |                                                                            |
+| ---------------------------------- | -------------------------------------------------------------------------- |
+| **Work the way you browse**        | Home launcher, tabs for chats and Work sessions, `⌘K` to jump anywhere     |
+| **Talk to the model you want**     | Local Ollama, Ollama Cloud, bundled providers, or any compatible endpoint  |
+| **Use the agent you subscribe to** | An installed Claude Code or Codex CLI, as a chat model, with no API key    |
+| **Give every task a workspace**    | Files, a terminal, a diff, and a live preview inside an isolated container |
+| **Bring your own context**         | Chat with PDF and plain-text documents by keyword or embeddings            |
+| **Turn answers into artifacts**    | Sandboxed HTML, SVG, JSON, code, and multi-file previews beside the chat   |
+| **Create with more than text**     | Image generation, provider-backed speech, and reusable personas            |
+| **Make the interface yours**       | Light and dark themes, adaptive accents, 25 locales including Arabic RTL   |
+| **Operate it your way**            | Local accounts and roles, optional OAuth, Docker, Kubernetes, npm, desktop |
+| **Extend without lock-in**         | Chat, embedding, image, and speech providers through the plugin layer      |
 
-## Local-first is a real boundary
-
-Libre WebUI ships without application telemetry or analytics. When you use Ollama locally, prompts and responses stay on the infrastructure you control. If you select a remote provider, that provider receives the requests you choose to send to it; Libre WebUI does not pretend otherwise.
-
-For shared and public deployments, the backend includes:
+## Freedom should survive success
 
-- Local accounts with bcrypt password hashing and JWT sessions
-- Admin and user roles, login/signup rate limits, and optional Cloudflare Turnstile
-- Optional GitHub and Hugging Face OAuth
-- Application-layer AES-256-GCM encryption for supported sensitive values such as credentials
-- Persistent SQLite storage, configurable data directories, and documented backup practices
+An AI interface can be self-hosted today and still become a gatekeeper
+tomorrow. Libre WebUI is built so your freedom does not depend on a company
+remaining benevolent.
 
-Application-layer encryption is not full-disk or end-to-end encryption. Use HTTPS, access-controlled storage, stable secrets, and disk encryption where your threat model requires them. Start with the [authentication guide](https://docs.librewebui.org/AUTHENTICATION) and [encryption guide](https://docs.librewebui.org/DATABASE_ENCRYPTION).
+Run it. Inspect it. Change it. Fork it. Connect it to something else. Or leave
+it behind. The code is Apache 2.0, local inference is the default path, remote
+providers are opt-in, and the project charter puts independence and inclusive
+participation into writing.
 
-## Models and providers without a single-provider worldview
+> **Open source is a license. Libre is an operating principle.**
 
-Ollama is the default local path, not a lock-in strategy. Libre WebUI also
-bundles an MLX LM plugin for native Apple Silicon inference, plus provider
-definitions spanning chat, images, embeddings, speech, and OpenAI-compatible
-services, including OpenAI, Anthropic, Google Gemini, Groq, Kimi Code by
-Moonshot AI, Mistral, OpenRouter, Hugging Face, GitHub Models, ComfyUI,
-ElevenLabs, Qwen3-TTS, and Kyutai TTS.
+Do not trust the pitch. Inspect the
+[license](https://github.com/libre-webui/libre-webui/blob/main/LICENSE),
+[charter](https://github.com/libre-webui/libre-webui/blob/main/CHARTER.md),
+[source](https://github.com/libre-webui/libre-webui), and
+[CI](https://github.com/libre-webui/libre-webui/actions).
+
+| What stays yours      | Libre WebUI's approach                                                  | Verify it                                                                  |
+| --------------------- | ----------------------------------------------------------------------- | -------------------------------------------------------------------------- |
+| **Data path**         | Self-hosted by default; remote inference only when you select it        | [Deployment docs](https://docs.librewebui.org/DOCKER)                      |
+| **Model choice**      | Local Ollama plus optional provider plugins and compatible endpoints    | [Plugin architecture](https://docs.librewebui.org/PLUGIN_ARCHITECTURE)     |
+| **Code**              | Use, modify, redistribute, and fork under Apache 2.0                    | [License](https://github.com/libre-webui/libre-webui/blob/main/LICENSE)    |
+| **Project direction** | The charter rejects funding control over roadmap, license, or community | [Charter](https://github.com/libre-webui/libre-webui/blob/main/CHARTER.md) |
+| **Participation**     | Everyone is welcome regardless of background, identity, or geography    | [Charter](https://github.com/libre-webui/libre-webui/blob/main/CHARTER.md) |
+| **Exit**              | Your installation and your fork do not need our permission              | [Source](https://github.com/libre-webui/libre-webui)                       |
 
-Credentials can come from deployment-wide environment variables or encrypted, user-scoped settings. Plugins can define static model fallbacks and use live discovery when a provider exposes a compatible model-list endpoint.
+## Local-first is a real boundary
 
-[Explore the plugin system →](https://docs.librewebui.org/PLUGIN_ARCHITECTURE)
+Libre WebUI ships without application telemetry or analytics. When you use
+Ollama locally, prompts and responses stay on infrastructure you control. If you
+select a remote provider, that provider receives the requests you choose to send
+it; Libre WebUI does not pretend otherwise.
+
+For shared and public deployments, the backend includes local accounts with
+bcrypt hashing and JWT sessions, admin and user roles, login and signup rate
+limits, optional Cloudflare Turnstile, optional GitHub and Hugging Face OAuth,
+application-layer AES-256-GCM encryption for sensitive values such as
+credentials, and persistent SQLite storage with configurable data directories.
+
+Application-layer encryption is not full-disk or end-to-end encryption. Use
+HTTPS, access-controlled storage, stable secrets, and disk encryption where your
+threat model requires them. Start with the
+[authentication guide](https://docs.librewebui.org/AUTHENTICATION) and
+[encryption guide](https://docs.librewebui.org/DATABASE_ENCRYPTION).
+
+## Models, providers, and agents
+
+Ollama is the default local path, not a lock-in strategy. Libre WebUI bundles an
+MLX LM plugin for native Apple Silicon inference plus provider definitions
+spanning chat, images, embeddings, and speech — OpenAI, Anthropic, Google
+Gemini, Groq, Kimi Code by Moonshot AI, Mistral, OpenRouter, Hugging Face,
+GitHub Models, ComfyUI, ElevenLabs, Qwen3-TTS, Kyutai TTS, and any
+OpenAI-compatible service.
+
+Credentials come from deployment-wide environment variables or encrypted,
+user-scoped settings. Plugins can declare static model lists and use live
+discovery when a provider exposes a compatible model-list endpoint.
+
+A coding agent you already pay for can answer in chat directly. If `claude` or
+`codex` is installed on the server, administrators see an **Agents** group in
+the model selector and can hold a normal conversation using the subscription
+that CLI is already signed in with — no API key. The CLI runs as the Libre WebUI
+server user and inherits that user's agent credentials, so treat it as
+equivalent to granting those agents shell access, and set
+`AGENT_CLI_MODELS_ENABLED=false` to turn it off.
+
+[Connect a provider →](https://docs.librewebui.org/PROVIDER_CONNECTIONS) ·
+[Use an installed agent →](https://docs.librewebui.org/AGENT_CLI_MODELS)
 
 ## Work: one persistent environment per task
 
-Work is Libre WebUI's built-in coding-agent surface. Choose **Work** beside
-**Chat**, select a tool-capable model, and describe what you want to build or
-change. Each task keeps its own conversation, exact provider route, files, tool
-activity, and preview state.
+Work gives a tool-capable model a real place to build. Describe what you want,
+and the task keeps its own conversation, provider route, files, tool activity,
+and preview — so returning to it a week later returns to exactly where it was.
 
 ```text
 Work task
    ├── durable conversation and run history in SQLite
    ├── dedicated Docker named volume mounted at /workspace
    ├── disposable, policy-checked container for commands
-   └── Files · Activity · Preview workspace pane
+   └── Files · Activity · Terminal · Preview workspace pane
 ```
 
-The model can list, read, write, and search task files, run bounded shell
-commands, and start or stop a browser preview. The workspace includes a
-light/dark syntax-highlighted editor, supported-file formatting, browser drafts,
-conflict-aware saves, and a draggable responsive split between conversation and
-workspace. Tasks remain in the main sidebar so returning to an older task also
-returns to its existing files and history.
-
-Active runs stream assistant text, provider-exposed reasoning, tool calls,
-results, usage, and state changes into the interface. Reasoning appears only
-when the selected provider returns it; Libre WebUI cannot reveal a model's
-hidden chain-of-thought. Server-owned worker skills teach the model the
-workspace boundary, efficient project discovery, safe editing, verification,
-and preview lifecycle without adding control files to the project.
-
-Work can use installed Ollama models, Ollama Cloud models, or configured
-completion/chat plugins through OpenAI-compatible, Anthropic, and Gemini tool
-adapters. Tool support is required. Selecting a remote route shows a dismissible
-per-user disclosure because an autonomous run may make multiple paid provider
-calls and may send conversation context and requested tool results—including
-file contents—to that service.
-
-The same configurable round budget applies to every provider route (48 by
-default). If it is exhausted, Work requests a final no-tools progress handoff
-and ends in the yellow **Needs input** state instead of reporting a false
-success or a raw provider-specific round-limit error. A follow-up run continues
-in the same durable workspace.
-
-Work is admin-only and never falls back to host command execution. Containers
-run as a non-root user with a read-only root filesystem, dropped capabilities,
-resource limits, and only the task volume mounted. Containers are still not
-virtual machines, current Work tasks have Docker bridge egress, and task volumes
-need their own backup and storage policy.
-
-[Read the Work workspace guide →](https://docs.librewebui.org/WORKSPACES)
-
-## Optional broader agents, with the runtime kept separate
-
-[Libre Claw](https://github.com/kroonen-ai/libre-claw) is an optional, admin-controlled local agent runtime. Libre WebUI provides the authenticated control surface for durable runs, timelines, approvals, schedules, usage, and configuration; Libre Claw owns the tools, memory, permission model, and execution.
-
-Libre Claw is separate from Work. Work is native to Libre WebUI and focused on
-one isolated project workspace; Libre Claw is the optional daemon for broader
-file, shell, git, browser, HTTP, web-search, MCP, memory, approval, and
-automation workflows.
+The model can read, write, move, delete, and search files, run bounded shell
+commands, and start or stop a browser preview. You get a syntax-highlighted
+editor, a red/green diff of what changed in the last turn, conflict-aware saves,
+and a real interactive terminal attached to the same container — under the same
+policy as the model's own tools, so it is a window into the sandbox rather than
+a way around it.
+
+Runs stream assistant text, tool calls, results, usage, and state changes live.
+Reasoning appears only when the provider returns it; Libre WebUI cannot reveal a
+model's hidden chain-of-thought. When a run exhausts its round budget it asks
+for a final progress handoff and ends in a yellow **Needs input** state instead
+of reporting false success — a follow-up run continues in the same workspace.
+
+Choosing a remote route shows a one-time disclosure, because an autonomous run
+can make many paid calls and can send conversation context and tool results,
+including file contents, to that provider. You should know that before a run
+starts, not after the bill.
+
+**On the sandbox.** Work is admin-only and never falls back to host execution.
+Containers run as a non-root user with a read-only root filesystem, dropped
+capabilities, `no-new-privileges`, resource limits with swap pinned to the
+memory cap, and only the task volume mounted. Networked tasks join a managed
+bridge with inter-container communication disabled. The whole policy is hashed
+into a container label and re-verified before reuse, so a container predating a
+hardening change is recreated rather than reused.
+
+Two honest limits: containers are not virtual machines, and Work tasks have
+outbound internet egress by design.
+
+A task can instead be bound to a real folder on your machine, when a deployment
+opts in with `WORK_HOST_WORKSPACES_ENABLED`. Requested paths are resolved
+through symlinks, checked against an allowlist of roots, and credential
+directories are refused — but it ships off, because pointing an agent at your
+real files is a deliberate narrowing of the sandbox and should be a decision you
+make on purpose.
+
+[Read the Work guide →](https://docs.librewebui.org/WORKSPACES)
+
+## Agents: broader automation, runtime kept separate
+
+The **Agents** section is the control surface for
+[Libre Claw](https://github.com/kroonen-ai/libre-claw), an optional
+admin-controlled local agent runtime. Libre WebUI provides the authenticated
+view of durable runs, timelines, approvals, schedules, usage, and configuration;
+Libre Claw owns the tools, memory, permission model, and execution.
+
+It is deliberately not Work. Work is native and focused on one isolated project
+workspace; Libre Claw is the optional daemon for broader file, shell, git,
+browser, HTTP, web-search, MCP, memory, and automation workflows. The WebUI
+backend never executes its shell or browser tools — it proxies authenticated
+admin actions to the separately running daemon, where approvals stay
+authoritative.
+
+[Read the integration guide →](https://docs.librewebui.org/LIBRE_CLAW_INTEGRATION)
+
+## How it fits together
 
 ```text
 Browser or desktop client
@@ -195,31 +250,30 @@ Browser or desktop client
           └────────► Libre Claw (optional agent runtime)
 ```
 
-The WebUI backend does not execute Libre Claw's shell or browser tools itself. It proxies authenticated admin actions to the separately running daemon, where approvals and safety rules remain authoritative.
-
-[Read the Libre Claw integration guide →](https://docs.librewebui.org/LIBRE_CLAW_INTEGRATION)
-
 ## Deploy on your terms
 
-| Path                | Command or link                                                         | Best for                                                              |
-| ------------------- | ----------------------------------------------------------------------- | --------------------------------------------------------------------- |
-| **npm**             | `npx libre-webui`                                                       | Fast local start; Work is available when local Docker is installed    |
-| **Docker + Ollama** | `docker compose up -d`                                                  | Persistent chat stack; Work enabled through the host Docker socket    |
-| **External Ollama** | `docker compose -f docker-compose.external-ollama.yml up -d`            | Existing Ollama; Work enabled the same way                            |
-| **NVIDIA Docker**   | `docker compose -f docker-compose.gpu.yml up -d`                        | Local GPU inference; Work enabled the same way                        |
-| **Kubernetes**      | `helm install libre-webui oci://ghcr.io/libre-webui/charts/libre-webui` | Cluster deployment; the current chart has no Work runtime driver      |
-| **Desktop client**  | [GitHub Releases](https://github.com/libre-webui/libre-webui/releases)  | Electron UI; Work runs wherever its separately managed backend runs   |
-| **Source**          | `npm install && npm run dev`                                            | Development; Work uses Docker available to the native backend process |
-
-Docker commands assume you have cloned this repository. Production deployments should set stable `JWT_SECRET` and `ENCRYPTION_KEY` values, persist the data directory, back up the database and key together, and terminate public traffic with HTTPS.
-
-Every Compose file above ships the Docker CLI and mounts the host Docker socket,
-so Work runs task containers as siblings of the Libre WebUI container with no
-extra flags. That socket grants root-equivalent control of the Docker host:
-treat every Libre WebUI administrator as a host administrator, and remove the
-`/var/run/docker.sock` mount from your Compose file if you do not want Work.
-
-On Linux the socket is owned by the `docker` group rather than root, so set the
+| Path                | Command or link                                                         | Best for                                                      |
+| ------------------- | ----------------------------------------------------------------------- | ------------------------------------------------------------- |
+| **npm**             | `npx libre-webui`                                                       | Fast local start; Work available when Docker is installed     |
+| **Docker + Ollama** | `docker compose up -d`                                                  | Persistent stack; Work enabled through the host Docker socket |
+| **External Ollama** | `docker compose -f docker-compose.external-ollama.yml up -d`            | An Ollama you already run                                     |
+| **NVIDIA Docker**   | `docker compose -f docker-compose.gpu.yml up -d`                        | Local GPU inference                                           |
+| **Kubernetes**      | `helm install libre-webui oci://ghcr.io/libre-webui/charts/libre-webui` | Cluster deployment; the chart has no Work runtime driver      |
+| **Desktop client**  | [GitHub Releases](https://github.com/libre-webui/libre-webui/releases)  | Electron UI over a separately managed backend                 |
+| **Source**          | `npm install && npm run dev`                                            | Development                                                   |
+
+Docker commands assume you have cloned this repository. Production deployments
+should set stable `JWT_SECRET` and `ENCRYPTION_KEY` values, persist the data
+directory, back up the database and key together, and terminate public traffic
+with HTTPS.
+
+> **The Docker socket is the security decision to understand.** Every Compose
+> file mounts `/var/run/docker.sock` so Work can run task containers as
+> siblings. That socket grants root-equivalent control of the host: treat every
+> Libre WebUI administrator as a host administrator, and remove the mount if you
+> do not want Work.
+
+On Linux the socket belongs to the `docker` group rather than root, so set the
 group id once in `.env`:
 
 ```bash
@@ -227,21 +281,23 @@ echo "DOCKER_GID=$(docker run --rm -v /var/run/docker.sock:/var/run/docker.sock
   alpine stat -c '%g' /var/run/docker.sock)" >> .env
 ```
 
-See the Work guide for the full security discussion.
-
-[Work](https://docs.librewebui.org/WORKSPACES) · [Docker](https://docs.librewebui.org/DOCKER) · [Kubernetes](https://docs.librewebui.org/KUBERNETES) · [Environment variables](https://docs.librewebui.org/ENVIRONMENT_VARIABLES) · [Hardware guide](https://docs.librewebui.org/HARDWARE_REQUIREMENTS)
-
-## Built from experience, for independence
-
-Libre WebUI is built around a hard-earned lesson: source availability alone does not guarantee shared power. Our response is constructive—compete by shipping better software and stronger commitments that anyone can inspect.
+## Documentation
 
-That means local-first defaults without forbidding the cloud, commercial support without changing the core license, and a community where identity, background, or geography never determines who belongs. Kroonen AI maintains Libre WebUI under the project's public charter, which accepts donations or grants only when they do not control the roadmap, license, or community.
+Full documentation lives at
+**[docs.librewebui.org](https://docs.librewebui.org)**.
 
-**Freedom should not be a phase in a startup's growth plan. It should be part of the architecture.**
+| Getting started                                                            | Going deeper                                                             | Operating it                                                               |
+| -------------------------------------------------------------------------- | ------------------------------------------------------------------------ | -------------------------------------------------------------------------- |
+| [Quick start](https://docs.librewebui.org/QUICK_START)                     | [Work workspaces](https://docs.librewebui.org/WORKSPACES)                | [Docker](https://docs.librewebui.org/DOCKER)                               |
+| [Working with models](https://docs.librewebui.org/WORKING_WITH_MODELS)     | [Provider connections](https://docs.librewebui.org/PROVIDER_CONNECTIONS) | [Kubernetes](https://docs.librewebui.org/KUBERNETES)                       |
+| [Keyboard shortcuts](https://docs.librewebui.org/KEYBOARD_SHORTCUTS)       | [Installed agent CLIs](https://docs.librewebui.org/AGENT_CLI_MODELS)     | [Environment variables](https://docs.librewebui.org/ENVIRONMENT_VARIABLES) |
+| [Hardware requirements](https://docs.librewebui.org/HARDWARE_REQUIREMENTS) | [Plugin architecture](https://docs.librewebui.org/PLUGIN_ARCHITECTURE)   | [Authentication](https://docs.librewebui.org/AUTHENTICATION)               |
+| [Troubleshooting](https://docs.librewebui.org/TROUBLESHOOTING)             | [Document chat](https://docs.librewebui.org/RAG_FEATURE)                 | [Database encryption](https://docs.librewebui.org/DATABASE_ENCRYPTION)     |
 
 ## Build with us
 
-Libre WebUI is for people who want excellent AI software and the freedom to outgrow its maintainers.
+Libre WebUI is for people who want excellent AI software and the freedom to
+outgrow its maintainers.
 
 1. Fork and clone the repository.
 2. Create a branch from `dev`.
@@ -249,26 +305,48 @@ Libre WebUI is for people who want excellent AI software and the freedom to outg
 4. Add tests and documentation with your change.
 5. Open a pull request against `dev`.
 
-You can also help by [reporting bugs](https://github.com/libre-webui/libre-webui/issues), improving documentation, translating the interface, or contributing provider definitions.
+You can also help by
+[reporting bugs](https://github.com/libre-webui/libre-webui/issues), improving
+documentation, translating the interface, or contributing provider definitions.
 
-All participation follows the [Contributor Covenant 2.1](https://www.contributor-covenant.org/version/2/1/code_of_conduct/) and the [Libre WebUI Community & Ethical Charter](https://github.com/libre-webui/libre-webui/blob/main/CHARTER.md). Report security issues privately to **security@kroonen.ai**.
+All participation follows the
+[Contributor Covenant 2.1](https://www.contributor-covenant.org/version/2/1/code_of_conduct/)
+and the
+[Libre WebUI Community & Ethical Charter](https://github.com/libre-webui/libre-webui/blob/main/CHARTER.md).
+Report security issues privately to **security@kroonen.ai**.
 
-## Release mirrors
+## Built from experience, for independence
 
-Every release tag is pushed explicitly to both
-[GitHub](https://github.com/libre-webui/libre-webui/releases) and the
-[Forgejo mirror](https://git.kroonen.ai/libre-webui/libre-webui/releases).
-After GitHub CI publishes a release, an idempotent mirror copies its release
-notes to Forgejo and adds named links to the GitHub-hosted desktop artifacts.
-This keeps both release indexes aligned without storing a second copy of every
-large binary. Maintainers can find the gated publication and backfill procedure
-in the [release automation guide](https://docs.librewebui.org/RELEASE_AUTOMATION).
+Libre WebUI is built around a hard-earned lesson: source availability alone does
+not guarantee shared power. Our response is constructive — compete by shipping
+better software and stronger commitments that anyone can inspect.
 
-## Stewardship and support
+That means local-first defaults without forbidding the cloud, commercial support
+without changing the core license, and a community where identity, background,
+or geography never determines who belongs. Kroonen AI maintains Libre WebUI
+under the project's public charter, which accepts donations or grants only when
+they do not control the roadmap, license, or community.
 
-[Kroonen AI](https://kroonen.ai) funds development and provides professional deployment, integration, security review, training, customization, and SLA-backed support. Commercial services do not change the Apache 2.0 license of the core project.
+**Freedom should not be a phase in a startup's growth plan. It should be part of
+the architecture.**
+
+## Stewardship and support
 
-For enterprise work, contact **enterprise@kroonen.ai**. To support independent development directly, [sponsor Libre WebUI](https://github.com/sponsors/libre-webui).
+[Kroonen AI](https://kroonen.ai) funds development and provides professional
+deployment, integration, security review, training, customization, and
+SLA-backed support. Commercial services do not change the Apache 2.0 license of
+the core project.
+
+For enterprise work, contact **enterprise@kroonen.ai**. To support independent
+development directly,
+[sponsor Libre WebUI](https://github.com/sponsors/libre-webui).
+
+Every release tag is published to both
+[GitHub](https://github.com/libre-webui/libre-webui/releases) and a
+[Forgejo mirror](https://git.kroonen.ai/libre-webui/libre-webui/releases), so
+neither index is a single point of failure. Maintainers can find the procedure
+in the
+[release automation guide](https://docs.librewebui.org/RELEASE_AUTOMATION).
 
 ## Project links
 

package/backend/dist/db.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"db.d.ts","sourceRoot":"","sources":["../src/db.ts"],"names":[],"mappings":"AAiBA,OAAO,QAAQ,MAAM,gBAAgB,CAAC;AA0BtC;;GAEG;AACH,wBAAgB,WAAW,IAAI,QAAQ,CAAC,QAAQ,CA0D/C;AAED;;GAEG;AACH,wBAAgB,eAAe,IAAI,QAAQ,CAAC,QAAQ,GAAG,IAAI,CAO1D;AAwhBD;;GAEG;AACH,wBAAgB,aAAa,IAAI,IAAI,CAMpC;AAED;;GAEG;AACH,wBAAgB,qBAAqB,IAAI,OAAO,CAa/C;AAGD,eAAe,WAAW,CAAC"}
+{"version":3,"file":"db.d.ts","sourceRoot":"","sources":["../src/db.ts"],"names":[],"mappings":"AAiBA,OAAO,QAAQ,MAAM,gBAAgB,CAAC;AA0BtC;;GAEG;AACH,wBAAgB,WAAW,IAAI,QAAQ,CAAC,QAAQ,CA0D/C;AAED;;GAEG;AACH,wBAAgB,eAAe,IAAI,QAAQ,CAAC,QAAQ,GAAG,IAAI,CAO1D;AAusBD;;GAEG;AACH,wBAAgB,aAAa,IAAI,IAAI,CAMpC;AAED;;GAEG;AACH,wBAAgB,qBAAqB,IAAI,OAAO,CAa/C;AAGD,eAAe,WAAW,CAAC"}