developers

Developer Resources

Everything on rashidazarang.com — essays, case studies, research, the CV, shipped products — is exposed for machine readers. Three surfaces, no keys, no tokens: an MCP server, a JSON API, and markdown content negotiation.

MCP server

Streamable HTTP at https://rashidazarang.com/mcp, also reachable at the well-known path /.well-known/mcp. Read-only, unauthenticated. Five tools: get_cv, search_essays, get_content, list_case_studies, list_projects. Add it to Claude, ChatGPT, Cursor, or any MCP-aware client:

{
  "mcpServers": {
    "rashidazarang": { "url": "https://rashidazarang.com/mcp" }
  }
}

A separate open-source MCP server for Airtable lives on npm as @rashidazarang/airtable-mcp.

Markdown content negotiation

Request any page with Accept: text/markdown and the response is the same URL as markdown — full text for essays and case studies, link indexes for listing pages. Responses vary on Accept, so caches keep the variants apart. Try: curl -H "Accept: text/markdown" https://rashidazarang.com/

JSON API

/api/context

Corpus structural overview: languages, topics, themes, concept index, reading paths.

/api/posts

Posts list as JSON — filter with ?lang=, ?topic=, ?audience=, ?content_format=.

/api/posts/<slug>

One post: frontmatter plus full markdown body.

/api/case-studies

Case studies list as JSON.

/api/case-studies/<slug>

One case study: metadata plus full markdown body.

/api/md/<slug>

Any essay or case study as raw text/markdown, no JSON wrapper.

/api/research

Research papers: metadata, abstract, DOI, rendered-HTML URL.

/api/books

Books metadata with URLs to rendered HTML.

/api/films

Film channel, episodes, studio, product videos.

/api/portfolio

Shipped platforms and open-source repos with star counts.

Indexes and feeds

/llms.txt — the agent-facing site index, including when-to-use guidance · /llms-full.txt — the entire corpus in one fetch · /sitemap.xml — every canonical URL · /feed.xml — RSS (English) · /feed/es.xml — RSS (Spanish).