> ## Documentation Index > Fetch the complete documentation index at: https://docs.rubixkube.ai/llms.txt > Use this file to discover all available pages before exploring further. # Kepler settings reference > Every Kepler setting, section by section: General, Appearance, Plan & Usage, Agents, Models, Tools & MCP, Gateway, Secrets, Permissions, Remote access, Memory, Watchers, Topology, Customize and Experiments. Open Settings with `Cmd+,` (`Ctrl+,` on Windows and Linux) or run `/settings` in the chat. Settings are grouped into **App**, **Agent** and **Labs**. This page walks the sections in the order they appear in the sidebar. Where a topic has its own page, this page stays short and links to it. ## App ### General | Setting | What it does | | ------------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Account | Who you are signed in as, and **Log out**. Signing out leaves your history, memory and settings on this device. | | About | The installed version and update status. | | Privacy | A statement of what leaves your machine: only calls to the model you chose and to your own tools. Conversation history and memory stay local. | | Dictation | Speak into the composer and Kepler types it out, using the operating system's own recognizer on your machine. macOS only. | | While you are away | Whether Kepler keeps running when you close the window, so watchers and scheduled work can still reach you. On macOS this needs **Allow in the Background** under **System Settings > General > Login Items**. Also: let watchers reach you on a channel, and run scheduled work on time. | ### Appearance | Setting | What it does | Default | | ----------------- | ------------------------------------------------------------------------------------------------------- | ---------------------- | | Appearance | Light, Dark, or System. System follows your OS. Applies to the app and the terminal. | System | | Theme | Pick a colour theme for the resolved mode, from Apple, Catppuccin, Dracula, Nord, Tokyo Night and more. | | | Accent | Optionally override the theme's highlight colour. | The theme's own accent | | Interface scale | Zoom the whole UI. | | | Tool-call density | How much detail each tool card shows in the chat. | | ### Plan & Usage Kepler runs locally and is free: no seats, no cloud, no limits. This section shows your plan and what this month has cost so far in model spend, updated as you work. Cloud plans add team sync, shared memory and hosted automations. ## Agent ### Agents | Setting | What it does | | ----------------- | ------------------------------------------------------------------------------------------------------------------------------------ | | Kepler | The agent you chat with. Toggle **thinking** to let it reason step by step before answering. | | Memory | The background worker that turns your conversations into lasting knowledge. A small, fast model handles this well, so keep it cheap. | | Topology | The worker that keeps your infrastructure map current. | | Shadow | A read-only watcher that pre-investigates incidents before you ask. Groundwork only in this release, not yet switched on. | | investigator | The built-in read-only specialist Kepler delegates to by default. | | Saved specialists | Specialists you created with `/agent`. See [Extend Kepler](/kepler/extend#specialist-agents). | ### Models Kepler runs on two models: a capable **main model** for reasoning and chat, and a fast **worker model** for the background lanes it spins up. Pick either from the composer's model picker or with `/model`. Whichever model you pick is the one that runs, every turn. Kepler never switches models on its own: if a call fails, it retries the same model or tells you. **Providers.** Enable a provider, add its key, and its models appear in the picker. Keys stay on this machine. | Provider | Notes | | ----------------- | ------------------------------------------------------------ | | Kepler | Hosted models, available once you sign in. No key to manage. | | Anthropic | Your own API key. | | OpenAI | Your own API key. | | Google Gemini API | Your own API key. | | Google Vertex AI | Your Google Cloud project credentials. | | xAI | Your own API key. | | AWS Bedrock | Your AWS credentials. | | Ollama | Local models on your machine. No key. | | OpenAI-compatible | Any endpoint that speaks the OpenAI API. Set the base URL. | **Context window.** | Setting | What it does | Options | | ------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------- | | Working window cap | How much of the model's window Kepler uses before compacting. Keeps cost and latency in check on large-window models. Flip **Max mode** in the composer to use the full window for one conversation. | 128K, 250K, 500K, 1M | | Auto-compact | Summarize older turns when the working window fills, so long sessions keep going. | On or off | | Compact at | How full the working window gets before compaction runs. | 75, 80, 82, 90 percent | You can also compact by hand at any time with `/compact`. ### Tools & MCP | Setting | What it does | | -------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------- | | Progressive tool discovery | Load MCP tools on demand instead of sending them all each turn. Keeps large integrations fast. | | Integrations | The catalog of integrations, their status, and custom MCP servers you add yourself. See [Extend Kepler](/kepler/extend#integrations-mcp-servers). | ### Gateway Connect WhatsApp, Slack, Telegram or Discord so you can carry on a session from your phone. Per channel: what it can do (Observe, Ask before changes, Act within bounds), default workspace, reply in channels on @mention, show typing, read receipts, and who can message. **Disconnect everything** turns off every channel and keeps the people you allowed. Reachable only while Kepler is running. See [Chat from your phone](/kepler/channels). ### Secrets Named credentials for MCP servers, stored in your operating system keychain and referenced as `${NAME}`. Each row shows which integrations use it. See [Extend Kepler](/kepler/extend#secrets). ### Permissions What you have allowed for good, your command rules, the always-blocked list, and the Files, Shell and MCP servers switches, each set to allow, ask or block. See [Permissions](/kepler/permissions). ### Remote access Switch **Work on other machines** on to let a thread work on a server, a cloud VM or a sandbox. Below it: the machines Kepler knows, each with Connect, Disconnect or Start and a per-machine set of limits; commands running in the background; and **New threads start on**. See [Remote access](/kepler/remote-access). ### Memory Import, export, reindex, rebuild the graph, wipe, and read shadow briefs. See [Memory](/kepler/memory#settings-memory). ### Watchers Every watcher across every session, with counts for watching, checks, alerts sent and ended, and the detail of each one. See [Watchers](/kepler/watchers). ### Topology | Setting | What it does | | ------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Scanning | On or paused. Kepler reads your clusters, hosts, cloud accounts and repositories on a schedule to keep its map current. Nothing leaves your machine, and it only looks at what you already have access to. | | Frequency | How often Kepler checks for changes. | | Sources | Each source it reads and its last status. | | Clear the map | Start the map over. | ### Customize | Setting | What it does | | ------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Signature | The name and emoji Kepler stamps on replies it sends on your channels, so its messages never read as yours. **Sign replies everywhere** is off by default because Telegram, Slack and Discord already show Kepler's own name. | | Skills | Create, import and edit skills. See [Extend Kepler](/kepler/extend#skills). | | Rules, Hooks | Listed as coming soon. Not available in this release. | ## Labs ### Experiments | Experiment | What it does | | ------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | BrowserV2 | Kepler drives your own Chrome instead of the embedded browser panel: real cookies, real extensions, real logged-in sessions. Install the extension, connect, then enable. | | Command line | Install or update the `kepler` command, which drives the running app from any terminal with the same agent and the same sessions. | Settings that talk to the agent show "Can't reach Kepler" if the app is still starting. Wait a moment and reopen the page. ## Where to go next The rules layer under the posture. What Kepler remembers and how to edit it. Gateway setup for each channel. When a setting does not take.