Skip to main content

Research & Case Studies

Research and case studies

What we have investigated, built and learned, written from evidence you can check. Every item states its evidence basis and its limits.

How to read the evidence labels
Client work
Based on a delivered client engagement, published with the client's permission.
Client work, anonymised
Based on a delivered client engagement. Identifying details are withheld.
Internal implementation
Based on systems PrajnaBytes built and runs itself. Not a client deployment.
Demonstration project
A demonstration or prototype built to test an approach. Not a client deployment.
Experimental research
Results from controlled tests. They may not hold in other environments.
Independent research
Research from publicly available evidence. The organisations discussed were not PrajnaBytes clients.
Desk research
Based on the cited public sources, with PrajnaBytes' interpretation marked as such.

Featured

Case studyIndependent research17 min read

AIIMS Delhi Website Ad Injection: A Historical Third-Party JavaScript Case Study

In December 2022 a visitor-counter script on the official AIIMS Delhi website was found to be placing third-party advertising at the top of the page. This case study reconstructs what happened from the original report and from archived copies of the script, separates verified facts from interpretation, and sets out what the episode teaches about third-party JavaScript.

  • Third-party JavaScript
  • Web security
  • Supply-chain risk
  • Content Security Policy

More publications

Technical insightInternal implementation4 min read

A strict Content Security Policy for a React website, without inline styles

How the PrajnaBytes website runs a Content Security Policy with no 'unsafe-inline' for scripts or styles, while still using React, UI libraries and dynamic effects: stylesheet classes, one per-request nonce, CSS custom properties set through the CSSOM, and self-hosted assets.

  • Content Security Policy
  • Web security
  • React

Technical insightInternal implementation4 min read

A readable XML sitemap without XSLT, before Chrome removes it

Chrome removes XSLT from stable releases in version 158 (17 November 2026), and with it the usual way of making an XML sitemap readable in a browser. CSS on XML is not being removed. This is how our sitemap uses a plain CSS stylesheet instead, and what that approach cannot do.

  • XML sitemaps
  • Search engine optimisation
  • Web standards

Facing a similar problem?

Tell us what you are dealing with. We will say honestly whether we can help, and how.

Talk to us
+91 99116 64994support@prajnabytes.com