/docs

jlopes.eu has a small read-only JSON API and a set of machine-readable files for agents.

Base URL and versioning

The current major version is v1, carried in the URL path:https://jlopes.eu/api/v1/. The unversioned/api/mcp/... paths are stable aliases of v1. Breaking changes only ever appear under a new version prefix; retired paths sendDeprecation and Sunset headers for at least 90 days before removal.

Endpoints

Full machine-readable spec: openapi.json(OpenAPI 3.1). Paths use trailing slashes; requests without one are redirected.

• GET /api/v1/mcp/profile/ — author profile: name, job title, skills, social links.
• GET /api/v1/mcp/content/ — every published post, project, and tool with slug, title, description, date, category, and tags.
• GET /api/v1/mcp/content/{slug}/ — one item by slug; posts include the first 3000 characters of the body.
• GET /api/v1/mcp/search/{q}/ — fuzzy search over titles, descriptions, and tags.

curl https://jlopes.eu/api/v1/mcp/search/react/
curl https://jlopes.eu/api/v1/mcp/content/minesweeper/

Errors

Errors are JSON, never HTML, with a real HTTP status code and a structured body:

{ "error": { "code": "not_found",
    "message": "No post or tool with slug \"nope\"",
    "hint": "List valid slugs at /api/mcp/content/" } }

Content as markdown

The homepage and every blog post negotiate markdown: request/ or /blog/{slug}/ withAccept: text/markdown to get markdown instead of HTML. Posts are also served raw at /blog/{slug}.md.

Machine-readable discovery

• /llms.txt — site overview for agents
• /llms-full.txt — full text of every post
• /openapi.json — OpenAPI 3.1 spec
• /.well-known/api-catalog — RFC 9727 linkset of everything on this page
• /.well-known/mcp/server-card.json — MCP server card
• /.well-known/agent-skills/index.json — public agent skills
• /sitemap-index.xml — sitemap
• /rss.xml — RSS feed

Questions? See /about for where to find me.