[click] to skip

Field notes · red team · AI security

Let's break the agents before advisories do.

I'm Saiprasad - I work in security, and I'm learning AI agent security in public. These are my lab notes: HTB writeups, red-team tooling, and a seven-part field manual on breaking and defending AI agents.

38 notes 1/7 published 0/7 read est. 2021

// Latest transmissions

Fresh from the lab notebook.

The newest notes - newest first, no filler. Everything here was broken, built, or both before it was written down.

Browse all 38 posts →

// Field manual - in progress

Building & securing AI agents, in public.

1 / 7 published 0 / 7 read

A seven-part field manual on AI agent security - architecture, hands-on labs, threat models, defense, and governance. Written as I learn it, so you can follow along or skip ahead.

01 published

Agent architecture basics

What an agent actually is - taxonomy, the execution loop, tools, memory, and the trust boundaries that matter.

Read the note →
02 draft

Before OWASP: Learn the Agent by Building It

Seven hands-on lab steps that teach you the agent by building one - before the frameworks arrive.

In the lab now
03 planned

OWASP systematization

Mapping what you built onto the OWASP agent threat model.

04 planned

Defending agents in production

Guardrails, evals, and the controls that survive contact with real users.

05 planned

MCP security deep dive

The Model Context Protocol under adversarial pressure.

06 planned

Governing agents

NIST AI RMF, ISO/IEC 42001, the EU AI Act - risk registers and control mapping.

07 planned

Assuring agents

Control testing, continuous monitoring, cloud deployment, incident response.

$ whoami

Saiprasad

handle
@At0micenergy
day job
security operations
after hours
breaking AI agents
method
break it, write it down

// Who writes this

I learn by breaking things - then writing down exactly how, so you don't have to guess.

Security practitioner by day, AI-agent tinkerer by night. This site is my public lab notebook: every writeup is something I actually ran, every series part is something I'm actively learning.

More about me →