This site talks to agents.

Every page is readable by people and by AI assistants. Agents can search, read and ask questions here through open standards, and every answer links back to the source.

What’s indexed right now

Published articles: 0
Topic guides: 6
Searchable passages: 42

In the browser: WebMCP

In browsers that support WebMCP (navigator.modelContext), the page registers these tools for the visitor’s own AI agent. Tools that change something only open a form; the visitor always confirms.

search_notesSearch the writing and topic guides; returns passages with deep links.
read_noteRead an article or topic guide as markdown.
open_noteTake the visitor to an article or topic.
summarize_noteSummarise one article, grounded in its text.
compare_notesCompare two articles.
list_topicsList the topic guides.
get_profileWho she is and how to reach her. Facts only.
show_subscribe_formOpen the newsletter form; the visitor enters their own email.
start_enquiryOpen the work-with-me form, optionally pre-filled; the visitor sends it.

From anywhere: the MCP server

https://kanikapahuja.com/mcp speaks the Model Context Protocol over Streamable HTTP (stateless, JSON-RPC over POST). All tools are read-only. Add it to an MCP client as a remote server.

search_notesKeyword search over passages, with deep links.
read_noteFull markdown of an article or topic guide.
related_notesArticles related to a given article.
list_topicsThe topic guides.
get_profileProfile facts and contact.
askA grounded answer with numbered sources, in English, Hindi or Hinglish.

curl -X POST https://kanikapahuja.com/mcp \
  -H "content-type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

Plain files for crawlers and agents

llms.txt
A short map of the site, with links to every article’s markdown.
llms-full.txt
The full text of every article in one file.
search-index.json
The passage-level search index the site itself uses.
Agent card
Experimental discovery document pointing to the MCP server.
Markdown twins
Add /index.md to any article URL for its plain markdown.
robots.txt
Search, assistant and training crawlers are all welcome by name.