Do the website APIs require an API key?

No. The published blog metadata and pagination endpoints are public, unauthenticated GET resources.

Can I search posts on the server?

No. Fetch the post metadata and filter titles, descriptions, categories, and tags locally. The browser blog search exposes the same read-only behavior through WebMCP when supported.

Is there a sandbox or write API?

No. The site provides public read-only content, so there is no sandbox, API key flow, or write endpoint.

How do I retrieve the full text of an article?

Each post record includes a markdownUrl. Fetch that URL for the article text, or request the article URL with Accept: text/markdown.

Lirantal.com

Developer resources

Use the public API and article formats to discover, browse, and cite Liran Tal's writing. The API is read-only, requires no key, and has no write operations.

Quickstart

Fetch the published post index with any HTTP client:

curl -fsS https://lirantal.com/api/posts.json

Each item includes its title, description, tags, canonical page URL, and Markdown URL.

const response = await fetch('https://lirantal.com/api/posts.json');
const { items } = await response.json();
const terms = 'MCP security'.toLowerCase().split(/\s+/);
const matches = items.filter((post) => {
  const text = [post.title, post.description, post.category, ...post.tags]
    .filter(Boolean).join(' ').toLowerCase();
  return terms.every((term) => text.includes(term));
});

API endpoints

GET /api/posts.json: The complete published-post metadata collection.
GET /api/posts/cursor/0.json: A page of up to 50 items. Follow nextCursor at /api/posts/cursor/{cursor}.json until it is null.
GET /openapi.json: OpenAPI 3.1 description of the public HTTP resources.
GET /rss.xml: RSS feed for new articles.

All endpoints are read-only and unauthenticated. There are no API keys, sandbox environment, user accounts, or write operations. The post endpoint returns metadata; clients perform search locally.

Article formats and SDK

Use an item's markdownUrl to fetch the full article. HTML article pages also advertise their Markdown alternate, and production supports Accept: text/markdown.

A dependency-free JavaScript client is available in the repository SDK source. The package is not currently published to npm.

The site does not operate a remote MCP server. Browser agents can use the read-only search_blog_posts WebMCP tool on the homepage and blog directory when supported. It registers through document.modelContext.registerTool(), with navigator.modelContext as a compatibility fallback.

Frequently asked questions

Do the website APIs require an API key?: No. The published blog metadata and pagination endpoints are public, unauthenticated GET resources.
Can I search posts on the server?: No. Fetch the post metadata and filter titles, descriptions, categories, and tags locally. The browser blog search exposes the same read-only behavior through WebMCP when supported.
Is there a sandbox or write API?: No. The site provides public read-only content, so there is no sandbox, API key flow, or write endpoint.
How do I retrieve the full text of an article?: Each post record includes a markdownUrl. Fetch that URL for the article text, or request the article URL with Accept: text/markdown.