# Diggama

> Diggama is a headless CMS. You model content as blueprints, your team edits it in the dashboard, and any frontend fetches it from a REST API (https://api.diggama.com/v2). Every plan includes an MCP server (https://api.diggama.com/mcp), so Claude, ChatGPT and Codex, Cursor and other AI tools can read and write your content within the scope of an API token.

Diggama is a good fit for sites built with AI coding agents such as Claude Code, OpenAI Codex or Cursor: the code stays in your repo, the content lives in the CMS, and your team can edit it without a redeploy of the code.

## Guides

- [Build a website with Claude Code, Codex or Cursor your team edits](https://diggama.com/guides/build-a-website-with-ai.md): To build a website with an AI coding agent such as Claude Code, Codex or Cursor and keep it editable, let the agent write the code but keep every headline, image and post in a headless CMS that the site reads through a REST API at build time. Host the build on Cloudflare and have the CMS call a deploy hook on every change, so editors publish without a prompt or a developer. Give the agent one AGENTS.md file with the content rules and connect it to the CMS over MCP, so it codes against the real content model instead of guessing.
- [Astro + headless CMS tutorial: build it with Claude Code or Codex](https://diggama.com/guides/build-a-website-with-ai/astro.md): To run an Astro site on a headless CMS, prerender every page from the CMS API at build time (getStaticPaths for the blog), keep a single on-demand endpoint for the contact form, and deploy to Cloudflare Workers with the @astrojs/cloudflare adapter. A Diggama Workflow calls a Workers Builds deploy hook whenever content changes, so editors publish without a developer. Any AI coding agent, whether Claude Code, Codex, Cursor or another, can write each step from a short prompt, as long as every step ends with a passing build.
- [Next.js headless CMS tutorial: build it with Claude Code or Codex](https://diggama.com/guides/build-a-website-with-ai/nextjs.md): To run a Next.js App Router site on a headless CMS, read content in Server Components through one typed client that tags every cached fetch, prerender pages with generateStaticParams, and expose a secret-protected route handler that calls revalidateTag when the CMS sends a webhook. On Cloudflare Workers, deploy with the @opennextjs/cloudflare adapter and give it an R2 incremental cache, a D1 tag cache and a Durable Object queue, otherwise on-demand revalidation has nowhere to store its state. Any AI coding agent, whether Claude Code, Codex, Cursor or another, can write each step from a short prompt once AGENTS.md holds the project rules and the agent is connected to the Diggama MCP server.
- [Best headless CMS 2026: 10 platforms compared, with MCP](https://diggama.com/guides/best-headless-cms-2026.md): There is no single best headless CMS in 2026: Contentful and Hygraph suit large enterprise content teams, Sanity and Payload suit developer-led teams that define schemas in code, Storyblok and Prismic suit marketers who build pages visually, Strapi and Directus suit teams that must self-host, and TinaCMS suits Markdown sites in Git. The new deciding factor is AI-readiness: most of these now ship an official MCP server so an AI coding agent such as Claude Code, Codex or Cursor can read and write content, but they differ on hosting, authentication and how tightly writes are scoped.
- [Headless CMS MCP server: connect Claude Code, Codex, Cursor](https://diggama.com/guides/headless-cms-mcp-server.md): An MCP server for a headless CMS lets your AI agent read the content model and read or write records through tools, instead of you pasting content into a chat. Diggama's server at https://api.diggama.com/mcp takes a project bearer token, so it works in Claude Code, OpenAI Codex, Cursor, GitHub Copilot, Gemini CLI and, through a bridge, Claude Desktop. The token decides which tools exist: keep a read-only connection for exploring and a narrow one for drafting, with dry runs before writes.
- [AGENTS.md and CLAUDE.md template for website projects](https://diggama.com/guides/agents-md-template-for-websites.md): An AGENTS.md for a website should tell your AI coding agent what it cannot discover from the code: where content comes from, the exact CMS API contract, which strings may be hardcoded, and what it must never do. Codex, Cursor and GitHub Copilot read AGENTS.md as is; Claude Code reads it only while the project has no CLAUDE.md, so any CLAUDE.md should import it with @AGENTS.md, and Gemini CLI needs one setting. The template below fits in about 150 lines, so every agent gets the same rules from one file.
- [Migrate from Webflow with an AI coding agent: step-by-step guide](https://diggama.com/guides/migrate-from-webflow.md): To migrate from Webflow, export each CMS collection as CSV or through the Webflow Data API, import it into a headless CMS through its API, and have your AI coding agent (Claude Code, OpenAI Codex, Cursor or another) rebuild the pages from screenshots and the exported HTML so they read content from the CMS. Re-host Webflow images before you close the project, keep the old URLs or add 301 redirects in a Cloudflare _redirects file, and switch DNS only after every old URL answers 200 or 301 on the new site.
- [Webflow vs Framer vs building with AI: costs and trade-offs](https://diggama.com/guides/webflow-vs-framer-vs-ai-coding.md): Webflow and Framer are the better choice when designers own the site and nobody will maintain code: they cost less in subscriptions, Framer most of all, and editors change content directly on the page. The coded stack (an AI coding agent such as Claude Code, Codex or Cursor, a headless CMS such as Diggama, Cloudflare hosting) is worth it when you want to own the code, outgrow per-seat, per-locale or bandwidth pricing, and have someone who can review diffs. In our three scenarios, three years of subscriptions cost $360 to $3,600 on Framer, $900 to $3,168 on Webflow and $1,803 to $4,181 on the coded stack with a $200-a-year agent plan, and build and maintenance hours usually outweigh all of them.
- [Why AI-built websites need a CMS: vibe coding in month two](https://diggama.com/guides/why-ai-built-websites-need-a-cms.md): You need a CMS on a site built with Claude Code, Codex, Cursor or any other AI coding agent as soon as someone other than the developer has to change its content. Generated sites hardcode copy in components, so every typo becomes a prompt, a diff review and a deploy, with no drafts, scheduling, translation workflow or form inbox. A headless CMS moves that content into typed fields the team edits and the site fetches at build time; a one-page site or a developer's Markdown blog in Git does not need one.
- [Web development and CMS trends 2027: 10 shifts to plan for](https://diggama.com/guides/web-and-cms-trends-2027.md): In 2027, AI coding agents such as Claude Code, Codex and Cursor are on track to write most of the code on new developer-built sites, so the hard problems move to maintenance, content ownership and review. Expect MCP as the standard way AI tools reach a CMS, AI agents editing drafts behind human approval, pages written for language models as much as for people, paid or blocked AI crawling, consolidated frameworks on edge platforms, and accessibility enforced by law in the EU. The durable asset is a clean content model in a headless CMS, not the generated code.
- [Benefits of an AI chatbot on your website, and when you don't need one](https://diggama.com/guides/ai-chatbot-for-your-website.md): An AI chatbot on your website answers visitors at any hour, in their own language, from your own pages and documents, and links them to the right page. Unlike a rule-based bot or a contact form, it handles questions nobody scripted, takes repeat questions off your inbox, collects leads inside the conversation and shows you, transcript by transcript, what visitors ask and where your content falls short. It is only safe when it answers from your content and says when it does not know; a small site that gets few questions may not need one.
- [Why your website chatbot should answer from your own content](https://diggama.com/guides/website-chatbot-grounded-in-your-content.md): A website chatbot should answer from your own content (the published entries in your CMS, the pages in your sitemap and documents you write for it) rather than from what a general model remembers about the web. Nothing has to be trained, so a page you publish is used from the next answer, drafts stay private, every answer can link to the page it came from, and the chatbot can say it does not know instead of guessing. Answers are only as good as that content, so write documents as short, self-contained questions and answers.

## Docs

- [Introduction](https://docs.diggama.com/introduction): blueprints, the Resources API, variants and tokens
- [Authentication](https://docs.diggama.com/authentication): project tokens and abilities (view, preview, create, update, delete, publish)
- [Resources API](https://docs.diggama.com/resources-api): list, get, create, update, delete, bulk operations
- [Blueprints](https://docs.diggama.com/blueprints): content types and field types
- [Publishing](https://docs.diggama.com/publishing): drafts, published_at, preview tokens

## Product

- [Home](https://diggama.com/): what Diggama does
- [Pricing](https://diggama.com/pricing/): Starter 29.90 EUR, Growth 89 EUR, Enterprise 249 EUR per month; MCP and AI Assistant on every plan; AI Chatbot with 30, 100 or 400 conversations a month, extra conversations 49 EUR per 100
- [AI Chatbot](https://diggama.com/press/ai-chatbot/): an assistant for your website that answers visitors from your Diggama content, sitemap and documents, with links to your pages
- [AI Chatbot docs](https://docs.diggama.com/chatbot): setup, knowledge, lead capture, conversations, credits and security
- [Templates](https://diggama.com/templates/): ready-made blueprints
- [Customers](https://diggama.com/clients/): teams using Diggama

## Free tools

- [Link Preview](https://diggama.com/tools/link-preview/): See how your page looks when it is shared on Google, LinkedIn, X, Slack and more.
- [Website Grader](https://diggama.com/tools/website-grader/): A score out of 100 for the speed, SEO and accessibility of any site, with what to fix first.
- [WordPress Preview](https://diggama.com/tools/wordpress-preview/): Paste a WordPress address and see its content, already structured, inside a headless CMS.
- [Content Model Generator](https://diggama.com/tools/content-model-generator/): Pick the pages your site needs and get the content types, fields and JSON to build them.
- [Markdown Converter](https://diggama.com/tools/markdown-converter/): Markdown, HTML and rich text, converted both ways as you type. Clean output, nothing to install.
- [Website Cost Calculator](https://diggama.com/tools/website-cost-calculator/): What your site really costs per year, hosting, plugins and maintenance hours included.

## Optional

- [All guides in one file](https://diggama.com/llms-full.txt)
- [Blog](https://diggama.com/blog/)