# agentregistry > This is the root documentation index. For detailed page listings, follow the section links below. ## agentregistry ### About agentregistry - [What is agentregistry?](https://aregistry.ai/docs/about/about/): Agentregistry is an open source, centralized AI artifact catalog that helps you build, package, … - [Core concepts](https://aregistry.ai/docs/about/concepts/): Agentregistry addresses the AI artifact management challenges that arise as development teams grow: … - [Architecture](https://aregistry.ai/docs/about/architecture/): Agentregistry is a single control plane that you install once and connect to one or more runtimes. … ### Setup - [Install on Kubernetes](https://aregistry.ai/docs/setup/kubernetes/): Use this guide to install agentregistry in a Kubernetes cluster by using Helm. This approach is … - [Connect a runtime](https://aregistry.ai/docs/setup/runtime/): Kagent is an open source, Kubernetes-native AI runtime. It extends Kubernetes with custom resources … - [Connect AI clients to the registry MCP server](https://aregistry.ai/docs/setup/mcp-client/): Agentregistry runs its own MCP server that exposes the artifact catalog as MCP tools on port 31313. … ### Agents #### Deploy - [kagent](https://aregistry.ai/docs/agents/deploy/kagent/): Before you begin Follow the Get started guide to install agentregistry. Connect the kagent runtime … ### MCP - [Local development](https://aregistry.ai/docs/mcp/local/llms.txt) ### Prompts - [Publish a prompt](https://aregistry.ai/docs/prompts/publish/): About prompts Prompts are reusable, versioned text strings that serve as system instructions for … ### Skills - [Create and publish a skill](https://aregistry.ai/docs/skills/publish/): About skills A skill is a reusable instruction set or slash command for AI agents that you can store … - [Pull skills](https://aregistry.ai/docs/skills/pull/): Before you begin Install agentregistry on Kubernetes. Create and publish a skill to the registry … ### Operations - [BYO PostgreSQL database](https://aregistry.ai/docs/operations/database/): By default, agentregistry deploys a bundled PostgreSQL database that stores the AI artifacts you … ### Reference #### CLI - [arctl](https://aregistry.ai/docs/reference/cli/arctl/): arctl is a CLI tool for managing agents, MCP servers, skills, and prompts. Global flags -h, --help … - [arctl apply](https://aregistry.ai/docs/reference/cli/arctl-apply/): Apply reads a YAML file (or stdin with -f -) containing one or more resource documents and applies … - [arctl build](https://aregistry.ai/docs/reference/cli/arctl-build/): Build the Docker image for a project created with arctl init. Reads arctl.yaml in the project … - [arctl completion](https://aregistry.ai/docs/reference/cli/arctl-completion/): Generate the autocompletion script for arctl for the specified shell. See each sub-command’s … - [arctl completion bash](https://aregistry.ai/docs/reference/cli/arctl-completion-bash/): Generate the autocompletion script for the bash shell. This script depends on the bash-completion … - [arctl completion fish](https://aregistry.ai/docs/reference/cli/arctl-completion-fish/): Generate the autocompletion script for the fish shell. To load completions in your current shell … - [arctl completion powershell](https://aregistry.ai/docs/reference/cli/arctl-completion-powershell/): To load completions in your current shell session: arctl completion powershell | Out-String | … - [arctl completion zsh](https://aregistry.ai/docs/reference/cli/arctl-completion-zsh/): Generate the autocompletion script for the zsh shell. If shell completion is not already enabled in … - [arctl configure](https://aregistry.ai/docs/reference/cli/arctl-configure/): Creates the .json configuration for each client, so it can connect to arctl. Global flags -h, … - [arctl configure claude-code](https://aregistry.ai/docs/reference/cli/arctl-configure-claude-code/): Write the MCP server entry that Claude Code reads, so it can reach this registry. The entry is … - [arctl configure cursor](https://aregistry.ai/docs/reference/cli/arctl-configure-cursor/): Write the MCP server entry that Cursor AI Editor reads, so it can reach this registry. The entry is … - [arctl configure kiro](https://aregistry.ai/docs/reference/cli/arctl-configure-kiro/): Write the MCP server entry that Kiro agentic IDE reads, so it can reach this registry. The entry is … - [arctl configure vscode](https://aregistry.ai/docs/reference/cli/arctl-configure-vscode/): Write the MCP server entry that Visual Studio Code reads, so it can reach this registry. The entry … - [arctl db](https://aregistry.ai/docs/reference/cli/arctl-db/): Work with the agent registry database directly, without starting the server. Covers schema … - [arctl db migrate](https://aregistry.ai/docs/reference/cli/arctl-db-migrate/): Apply, roll back, and inspect database migrations independently of server startup. Reads … - [arctl db migrate down](https://aregistry.ai/docs/reference/cli/arctl-db-migrate-down/): Undo the last N applied migrations, newest first, by running each migration’s .down.sql. … - [arctl db migrate force](https://aregistry.ai/docs/reference/cli/arctl-db-migrate-force/): Used to reconcile the selected source’s schema_migrations table after manual remediation. The … - [arctl db migrate goto](https://aregistry.ai/docs/reference/cli/arctl-db-migrate-goto/): Move the selected source’s schema to version V (forward or backward). V=0 is the special … - [arctl db migrate status](https://aregistry.ai/docs/reference/cli/arctl-db-migrate-status/): Report how many migrations are applied and how many are still pending, without changing anything. … - [arctl db migrate up](https://aregistry.ai/docs/reference/cli/arctl-db-migrate-up/): Applies pending migrations for every registered source in registration order. Per source, the … - [arctl db migrate version](https://aregistry.ai/docs/reference/cli/arctl-db-migrate-version/): Print the highest applied migration version. For a single registered source the value is on one … - [arctl delete](https://aregistry.ai/docs/reference/cli/arctl-delete/): Delete a registry resource by type and name, or from a YAML file. File mode (declarative): reads … - [arctl get](https://aregistry.ai/docs/reference/cli/arctl-get/): List every resource of a type, or fetch a single one by name. Supported types: agents, mcps, … - [arctl init](https://aregistry.ai/docs/reference/cli/arctl-init/): Scaffold a new declarative resource project: an agent, MCP server, skill, or prompt. The generated … - [arctl init agent](https://aregistry.ai/docs/reference/cli/arctl-init-agent/): Scaffold a new agent project with declarative YAML and framework stubs. Picks a framework + … - [arctl init mcp](https://aregistry.ai/docs/reference/cli/arctl-init-mcp/): Scaffold a new MCP server project with declarative YAML and framework stubs. NAME must be DNS-1123 … - [arctl init prompt](https://aregistry.ai/docs/reference/cli/arctl-init-prompt/): Create a new <name>.yaml in the current directory using the ar.dev/v1alpha1 declarative … - [arctl init skill](https://aregistry.ai/docs/reference/cli/arctl-init-skill/): Scaffold a new skill project with declarative YAML and source stubs. Creates a project directory … - [arctl pull](https://aregistry.ai/docs/reference/cli/arctl-pull/): Fetch a registry resource’s source repository to a local directory. Supported types: agent, … - [arctl run](https://aregistry.ai/docs/reference/cli/arctl-run/): Run the agent or MCP server defined by the declarative YAML in the project directory (defaults to … - [arctl version](https://aregistry.ai/docs/reference/cli/arctl-version/): Displays the version, git commit, and build date of the arctl CLI, and of the registry server when … - [arctl wait](https://aregistry.ai/docs/reference/cli/arctl-wait/): Wait for a registry resource to reach a target state, polling until it gets there or the timeout … --- Generated on 2026-09-02 22:34:30 UTC Site: https://aregistry.ai