# spaceduck

## Docs

- [Memory](https://docs.spaceduck.ai/concepts/memory.md): How spaceduck remembers, deduplicates, and self-corrects across conversations.
- [Scheduler](https://docs.spaceduck.ai/concepts/scheduler.md): How spaceduck runs tasks on its own — on a timer, on a cron schedule, or in response to events — with budget limits that prevent surprise bills.
- [Security](https://docs.spaceduck.ai/concepts/security.md): How spaceduck prevents injection attacks, isolates memory, enforces budgets, and sandboxes skills.
- [Skills](https://docs.spaceduck.ai/concepts/skills.md): How spaceduck loads, validates, and sandboxes community skills using SKILL.md files.
- [Introduction](https://docs.spaceduck.ai/index.md): Spaceduck is a local-first AI assistant with persistent memory. It remembers what you've said, acts on your behalf with real tools, and runs entirely on your machine.
- [Docker](https://docs.spaceduck.ai/install/docker.md): Run Spaceduck in Docker with persistent storage using a single container for the gateway and CLI.
- [From Source](https://docs.spaceduck.ai/install/from-source.md): Clone and build Spaceduck from source for development or manual control.
- [Installer](https://docs.spaceduck.ai/install/installer.md): Install Spaceduck with a single command.
- [CLI](https://docs.spaceduck.ai/platforms/cli.md): Manage your Spaceduck gateway from the terminal — check status, update config, and manage secrets.
- [Desktop App](https://docs.spaceduck.ai/platforms/desktop.md): Run Spaceduck as a native desktop application using Tauri v2, with system-wide Fn key dictation.
- [Gateway](https://docs.spaceduck.ai/platforms/gateway.md): The Spaceduck gateway is the core server that handles chat, memory, tools, and configuration.
- [AWS Bedrock](https://docs.spaceduck.ai/providers/bedrock.md): Connect Spaceduck to Amazon Bedrock for access to Nova, Claude, and Titan embeddings.
- [Google Gemini](https://docs.spaceduck.ai/providers/gemini.md): Connect Spaceduck to Google AI for Gemini chat models and text embeddings.
- [llama.cpp](https://docs.spaceduck.ai/providers/llamacpp.md): Run Spaceduck with llama-server for fully local chat and embeddings.
- [LM Studio](https://docs.spaceduck.ai/providers/lmstudio.md): Run Spaceduck with LM Studio for a GUI-based local model experience with chat and embeddings.
- [OpenRouter](https://docs.spaceduck.ai/providers/openrouter.md): Access hundreds of AI models through a single API key with OpenRouter.
- [Model Providers](https://docs.spaceduck.ai/providers/overview.md): Spaceduck supports both local and cloud model providers. Choose what fits your setup, or mix them.
- [Quickstart](https://docs.spaceduck.ai/quickstart.md): Get Spaceduck running and send your first message in under 5 minutes.
- [Common Errors](https://docs.spaceduck.ai/reference/common-errors.md): Solutions for the most frequent issues when running Spaceduck.
- [Browser](https://docs.spaceduck.ai/tools/browser.md): Headless browser automation using Playwright with accessibility-snapshot element references.
- [Charts](https://docs.spaceduck.ai/tools/charts.md): Render interactive bar, line, area, and pie charts directly in the conversation.
- [Document Scanning](https://docs.spaceduck.ai/tools/document-scanning.md): Convert PDFs to markdown using Marker — upload via the chat UI and the assistant processes them automatically.
- [Tools](https://docs.spaceduck.ai/tools/overview.md): Spaceduck comes with built-in tools that let the assistant browse the web, fetch pages, search, and process documents.
- [Web Fetch](https://docs.spaceduck.ai/tools/web-fetch.md): Fetch any URL and convert HTML to clean text — a lightweight alternative to the browser tool.
- [Web Search](https://docs.spaceduck.ai/tools/web-search.md): Search the web via Brave or SearXNG, and get AI-synthesized answers via Perplexity Sonar.

## OpenAPI Specs

- [openapi](https://docs.spaceduck.ai/api-reference/openapi.json)