/categories/research
Research & Findings
Firsthand findings, case studies, validation, and engineering retrospectives.
7 posts
- My Honeypot Dashboard Was Lying to Me for Two Weeks (and I Wrote It)273,469 events. 2,654 IPs. 54 countries. And a dashboard that confidently told me '7 countries' and '200 sessions' and a graph that quietly hid 95% of the data. The story of shipping v1.9 of ShardLure: not new attacks, but every way a tool misrepresents scale — and how I caught it.16 min read
- From Search Results to Two CERT-In RecognitionsA restrained account of using indexed government-domain pages as reconnaissance leads, preserving evidence, and separating two CERT-In recognitions from unsupported conclusions.5 min read
- From an Exposed Service to an Unresolved SQL Injection LeadAn evidence-bounded reconstruction of a Shodan lead that historical notes associated with SQL injection, but surviving artifacts cannot independently confirm.6 min read
- Five Findings That Looked Small Until I Followed the StateField notes on response handling, OTP controls, wallet state, and upload boundaries—and on the evidence needed to distinguish client behavior from server-side impact.8 min read
- I Thought I Found a Subdomain Takeover. I Had Not.A false-positive investigation showing why suspicious DNS and verification records did not establish a claimable third-party resource.7 min read
- When Object Authorization and Output Encoding Fail TogetherA bounded analysis of object access and three rendering contexts, including what one controlled callback proved—and what the apparent chain did not.8 min read
- The Invoice Number Changed. Did the Authorization Decision?A historical invoice authorization concern recast as a controlled two-account method for testing and preventing broken object-level authorization.6 min read