You probably think search is search. Type a word, get results. Done.<\/p>\n
That assumption is exactly what makes EmDash’s search so easy to underestimate. It’s not a better version of the same thing. It’s built on a completely different foundation that changes what search can actually do.<\/p>\n
EmDash<\/a> stores content as structured JSON \u2014 a format called portable text \u2014 instead of the HTML blobs WordPress dumps into WordPress search can’t do that. It searches a mess of concatenated HTML and hopes for the best.<\/p>\n Here’s what that means for anyone actually running a site.<\/p>\n EmDash runs on a D1 database (Cloudflare’s SQLite-based serverless database) with Astro handling the frontend. Every piece of content \u2014 every post, page, project, or custom content type \u2014 is stored as structured fields, not as a single blob of markup.<\/p>\n When you<\/a> create a content type in EmDash, you define each field individually. A “Project” content type might have a Search operates on these individual fields, not on a rendered HTML string.<\/p>\n That means you can tell EmDash: “Only search the title and description fields, not the URL field.” Or: “Boost results where the search term appears in the title over results where it appears in the body.” Or: “Exclude draft content from search entirely.”<\/p>\n These are not post-install plugin features. They’re built into how the CMS handles data.<\/p>\n The default EmDash installation includes search functionality that covers the basics competently and a few things WordPress requires plugins for.<\/p>\n The table makes it obvious: EmDash’s search advantage is architectural, not feature-bloated. It doesn’t add more stuff \u2014 it stores stuff better from the start.<\/p>\n Customization happens at two levels: the content type definition and the query configuration.<\/p>\n When you create or edit a content type in EmDash, each field has a “searchable” toggle. You access this through the admin panel under Content Types<\/strong>.<\/p>\n Fields marked searchable get included in the full-text index. Fields you leave off \u2014 like internal notes, URLs, or system IDs \u2014 are excluded. This is a per-field decision, not a global setting.<\/p>\n Content types also have an SEO Enabled<\/strong> toggle. When you enable this:<\/p>\n This is how EmDash handles what WordPress requires Yoast or Rank Math to do. It’s not a plugin. It’s a checkbox on the content type definition.<\/p>\n EmDash ships with an MCP (Model Context Protocol) server. This isn’t a search feature for end users \u2014 it’s for developers and AI agents<\/a>.<\/p>\n You can connect Claude, Cursor, or any MCP-compatible tool to your EmDash instance and run structured search queries directly. For example:<\/p>\n “Find all posts where the description field contains ‘TypeScript’ and the status is published, then return the title and URL.”<\/p>\n The MCP server translates that into the correct database query against the structured fields. No parsing HTML. No regex scraping content. The agent gets exactly the fields it asked for.<\/p>\n This is the AI-native search angle that WordPress is still trying to retrofit.<\/p>\n The default EmDash starter templates (Blog, Marketing Site, Portfolio) include a basic search interface. If you want something different, you build it in Astro \u2014 which is the frontend framework EmDash uses.<\/p>\npost_content<\/code>codecodecodecode. That single architectural decision means EmDash’s search doesn’t just match keywords. It understands the shape<\/em> of your content. It knows which field is a title, which is a body paragraph, which is a metadata tag, and which is an image caption. It can search each one independently or in combination.<\/p>\nThe Architecture That Makes Search Smarter<\/h2>\n
client_name<\/code>codecodecodecode field, a year<\/code>codecodecodecode field, a description<\/code>codecodecodecode field (using portable text for rich formatting), and a url<\/code>codecodecodecode field. Each one is a separate column in the database, with its own data type and index.<\/p>\nWhat EmDash Search Ships With Out of the Box<\/h2>\n
\n\n
\n \nFeature<\/th>\n EmDash (built-in)<\/th>\n WordPress (built-in)<\/th>\n WordPress (with plugin)<\/th>\n<\/tr>\n<\/thead>\n \n Field-level search scoping<\/td>\n Yes \u2014 per-field configuration in content types<\/td>\n No \u2014 searches entire post_content<\/code>codecodecodecode blob<\/td>\nYes \u2014 via Relevanssi or SearchWP<\/td>\n<\/tr>\n \n SEO metadata as searchable fields<\/td>\n Yes \u2014 title, description, canonical included by default<\/td>\n No \u2014 requires Yoast or similar<\/td>\n Yes<\/td>\n<\/tr>\n \n Content type filtering<\/td>\n Yes \u2014 native content type system<\/td>\n No \u2014 only posts and pages<\/td>\n Yes \u2014 via CPT plugin<\/td>\n<\/tr>\n \n Draft\/status-aware search<\/td>\n Yes \u2014 respects publication status<\/td>\n Yes \u2014 respects status<\/td>\n Same<\/td>\n<\/tr>\n \n Search indexing for custom fields<\/td>\n Yes \u2014 fields are database columns<\/td>\n No \u2014 custom fields are in wp_postmeta<\/code>codecodecodecode<\/td>\nYes \u2014 via ACF + search plugin<\/td>\n<\/tr>\n \n Built-in sitemap integration<\/td>\n Yes \u2014 content types can be marked SEO-enabled<\/td>\n No \u2014 requires plugin<\/td>\n Yes<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n How to Customize Search in EmDash<\/h2>\n
Step 1: Configure Which Fields Are Searchable<\/h3>\n
\n
Step 2: Mark Content Types for SEO<\/h3>\n
\n
Step 3: Use the Built-in MCP Server for Search Queries<\/h3>\n
Step 4: Customize the Frontend Search Experience<\/h3>\n