Agent guides
Connecting AI agents, building modules, kits and themes with them, and the rules they work under.
Publish a package to the registry
How a module, kit or theme reaches the registry on spillwaycms.com (or any Spillway instance running with REGISTRY_HOST=1), from a terminal or from an agent. The registry contract is in packages/shared/src/index.ts under Registry; the operator's side is in modules.md.
Build a theme with an agent
A theme in Spillway is a kit of tokens, blocks (shared blocks and page templates built from the core blocks) and page templates, with screenshots, published to the registry with "kind": "theme". Themes do not ship code: the look comes from the theme tokens the server emits as CSS custom…
Build a kit with an agent
A kit is a JSON manifest that installs a site's structure: content types, vocabularies with terms, collections, forms, menus, starter pages and sample content. It is the quickest thing to build with an agent, because the format is declarative and the server validates it. Read kits.md first; the…
Build a module with an agent
A recipe for having Claude Code, Cursor or any coding agent write a Spillway module: the prompt to paste, what the agent should produce, and the checks to run before you trust it. Read modules.md once yourself; the agent will read it too.
Rules for agents
What an AI agent working with Spillway must respect, whether it edits content on a site through MCP, writes code in this repository, or publishes a package to the registry. These rules are also encoded in the server: tokens cannot grow permissions, the AI layer has no publish path, and the…
Connect an agent to Spillway
Spillway treats AI agents as first-class readers and builders. An agent connects to a running site with an API token over the MCP server (POST /mcp) or the plain REST API, reads the content model, drafts content, runs the pre-publish checks and hands its work to a person. Every action is…