# Docs - [TulipFarm](/docs): The AI-native business operating system — install it, give it a soul, put it to work. - [Install (self-host)](/docs/installation): One-line install of the TulipFarm OCI stack on Linux, macOS, or Windows (WSL2). - [Get started](/docs/getting-started): Install TulipFarm, run it, log in, and define your first resource type. - Concepts - [Concepts](/docs/concepts): How TulipFarm's parts fit together — soul, resources, agents, skills, knowledge, LLM tiers, and secrets. - [Building by chat](/docs/concepts/building): How you build your system in TulipFarm — you describe what you want, and agents build it. - [Soul](/docs/concepts/soul): The git-backed configuration store that defines a TulipFarm instance. - [First-run setup and boot modes](/docs/concepts/setup): How TulipFarm decides whether to open the setup wizard or seed from environment variables. - [Resources](/docs/concepts/resources): Schema-defined data types and the records that fill them. - [Agents](/docs/concepts/agents): The actors that run turns, each defined by an AGENT.md file. - [Skills](/docs/concepts/skills): Reusable instructions agents load on demand, audited before they land. - [Routines](/docs/concepts/routines): Durable, repeatable automations that run without you — and pause for you when it matters. - [Integrations](/docs/concepts/integrations): MCP servers that give agents access to external tools, connected through the soul. - [Knowledge](/docs/concepts/knowledge): The searchable, interlinked corpus your agents and your team draw on. - [LLM tiers](/docs/concepts/llm): How TulipFarm chooses a model and falls back across providers. - [Secrets](/docs/concepts/secrets): Encrypted storage for the credentials your instance needs. - Guides - **Setup** - [Configure LLM providers](/docs/guides/configure-llm-providers): Add a provider's credentials, then point your tiers at its models. - [Store a secret](/docs/guides/store-a-secret): Add and remove encrypted secrets through the web UI. - [Issue an API token](/docs/guides/issue-an-api-token): Create a bearer token for programmatic access to your instance. - [Deploy without the setup wizard](/docs/guides/deploy-headless): Seed the first admin and LLM key from environment variables for automated and container-platform deployments. - **Operations** - [Run operational reviews](/docs/guides/operations): Review access, retention, audit health, and recovery evidence. - [Respond to an incident](/docs/guides/incidents): Contain mutations, preserve evidence, recover, and close an incident. - **Resources** - [Define a resource type](/docs/guides/define-a-resource-type): Create a new resource type by describing it to the assistant. - [Work with records](/docs/guides/work-with-records): Create, edit, and delete records of an existing resource type in the web UI. - **Skills** - [Install a skill](/docs/guides/install-a-skill): Scan a git repository for skills, audit them, and confirm the install. - **Agents** - [Create an agent](/docs/guides/create-an-agent): Describe an agent to the assistant and have it built for you. - **Routines** - [Build and run a routine](/docs/guides/routines): Describe an automation to the assistant, run it, watch it, and approve it when it pauses. - **Integrations** - [Connect an integration](/docs/guides/connect-an-integration): Install an MCP integration from the marketplace and connect it so agents can use its tools. - **Knowledge** - [Use the knowledge wiki](/docs/guides/use-the-knowledge-wiki): Create a space, write linked pages, and navigate them through backlinks, the graph, and search. - [Cite knowledge in chat](/docs/guides/cite-knowledge-in-chat): Ground an agent's answer in your knowledge base — read its citations and pin a specific page. - Reference - [Environment variables](/docs/reference/environment-variables): Every variable TulipFarm reads, with defaults and descriptions. - [Commands](/docs/reference/commands): The scripts that set up, run, and check a TulipFarm checkout. - Security - [Encryption](/docs/security/encryption): How TulipFarm encrypts the secrets your instance stores. - [Hashing](/docs/security/hashing): How TulipFarm hashes passwords and API tokens. - [Privacy operations](/docs/security/privacy): Product privacy controls and the deployment decisions that remain with operators. - [Control evidence](/docs/security/control-evidence): Implemented security and privacy controls, operator responsibilities, and evidence.