Field notes on
AI-native security.
How we build the scanner, what we find in the wild, and how to harden modern AI stacks. Long-form, technical, no fluff.
What is prompt injection? Types, examples, and how to test for it.
Prompt injection is an attack where untrusted text makes an LLM follow an attacker's instructions instead of its owner's. Direct vs indirect, real examples, defences, and safe testing.
Read the post →OWASP Top 10 for LLM Applications (2025) explained, with a test for each.
A plain-English walkthrough of all ten OWASP LLM risks, what each looks like in a real product, and the external test that reveals it.
Read the post →Prompt injection vs jailbreaking: what is the difference?
Prompt injection hijacks an application by hiding instructions in data the model reads. Jailbreaking talks a model out of its own safety rules. Why the distinction matters for testing.
Read the post →What is an AI security scanner? How it differs from a vulnerability scanner.
An AI security scanner tests the AI surface of an application, LLM endpoints, agents, and retrieval, alongside the web surface. What it checks, how it differs from classic scanners, and what to look for.
Read the post →How to check if a website is secure: a 10-minute external checklist.
Ten checks you can run from a browser and a terminal to tell whether a website is secure: TLS, security headers, exposed files, admin paths, cookies, third-party scripts, source maps, and subdomains.
Read the post →HTTP security headers explained: CSP, HSTS, X-Frame-Options, and what to set in 2026.
What each HTTP security header does, the value to set, and the mistakes scanners flag most. Includes a copy-paste baseline for Nginx, Next.js, and Express.
Read the post →What is external attack surface management (EASM)? EASM vs vulnerability scanning.
External attack surface management continuously discovers every internet-facing asset you own and tests it. How EASM differs from vulnerability scanning and pentesting, and how to start this week.
Read the post →Exposed .env files, source maps, and API keys in JavaScript: how leaks happen and how to find them.
The three most common ways secrets reach the public web: .env files served by the web server, published source maps, and keys bundled into front-end JavaScript. How to find and fix each.
Read the post →DPDP Act compliance checklist for websites: what an external scan can verify.
A practical checklist for India's Digital Personal Data Protection Act 2023 covering notice, consent, grievance redressal, security safeguards, children's data, and breach readiness, with the parts you can verify from outside.
Read the post →GDPR, DPDP, SOC 2 — the minimum scan you need before an audit.
A control-mapped checklist of what an external security scan can evidence for GDPR, India's DPDP Act, and SOC 2, and what it cannot. Run it before the auditor does.
Read the post →Scoring real breach risk: likelihood × impact, not raw CVE counts.
How Pythorix ranks findings by likelihood times impact, why unverified exploit claims cap at Medium, and the false-positive lessons behind the severity gate.
Read the post →Five surfaces every AI product team forgets to scan.
Stale staging hosts, debug endpoints, public buckets, source maps, and admin paths: the five most-missed exposures we find on AI product companies, and how to look for them.
Read the post →How we probe AI assistants for prompt injection — without breaking them.
How Pythorix tests production LLM endpoints for prompt injection safely: discovery first, fixed request budgets, canary tokens, and verification before severity.
Read the post →Want to be notified when we publish? Get on the list →
Read it. Then scan your stack.
Free public scan. No card. Live in 2 minutes.