API & agent documentation
Everything published on this site is also available as machine-readable data — over a plain JSON API, over MCP, and as Markdown. No API key, no registration, no rate-limit tiers: just get in touch if you need something that isn't here.
JSON API
Read-only, unauthenticated, described by
an OpenAPI 3.1 document. Copy is returned in the active
locale — append ?lang=nl for Dutch. Responses are wrapped in a data key.
| Endpoint | Returns |
|---|---|
GET /api/v1/services |
Every service offered, with its page URL and stack. |
GET /api/v1/themes |
Themes and templates: status, pricing model, price, demo URL. |
GET /api/v1/themes/{theme} |
A single theme by key, e.g. elephantro. |
GET /api/v1/blog |
Published articles, newest first. Supports ?q= and ?limit=.
|
GET /api/v1/blog/{slug} |
A single article including its original Markdown body. |
GET /up |
Health check. |
MCP server
A Model Context Protocol server is available over streamable HTTP at
https://eliancodes.com/mcp. It exposes four read-only tools —
list-services, list-themes, search-blog-posts and
get-blog-post — backed by the same data as the JSON API. Its
server card describes it for
automated discovery.
Markdown instead of HTML
Send Accept: text/markdown to any page on this site and you get a Markdown
representation of it back instead of HTML, with an x-markdown-tokens header
estimating its size. Articles are served as their original Markdown source.
curl -H 'Accept: text/markdown' https://eliancodes.com/blogDiscovery documents
Agents that don't know any of the above can start from the root and follow links. Every HTML and
JSON response carries Link headers (RFC 8288) pointing at the catalog, the OpenAPI
description, this page, the health endpoint and the Markdown alternate.
/.well-known/api-catalog— RFC 9727 linkset tying the API to its description, docs and status endpoint./.well-known/agent-skills/index.json— Agent Skills, with SHA-256 digests./.well-known/mcp/server-card.json— MCP server card./auth.md— how authentication works here (short version: it doesn't, everything is public)./sitemap.xmland/robots.txt— the latter declares content signals for AI training, search and AI input.
Terms
Use of the API is covered by the terms of service. Please attribute and link back when you reuse content, and keep request volumes reasonable — API routes are rate limited per IP.