I got tired of juggling 15 browser tabs during engagements, so I built this

#security #opensource #github #cybersecurity

There's a specific kind of frustration that happens mid-engagement.

You know what you're looking at. You've seen SeImpersonatePrivilege
before. You know there's a Potato attack that fits here. But the
exact tool for this Windows version, the right flags, whether to
reach for GodPotato or PrintSpoofer in this situation - it's not
sitting cleanly in your head right now.

So you open HackTricks. Then a blog post someone bookmarked on
Reddit three months ago. Then a GitHub repo you're not sure is
still maintained. Then a gist with no date on it. Twenty minutes
pass. The shell is still waiting.

The information was never the problem. It exists everywhere.
The problem is that it's scattered, inconsistent, written at
different depths for different audiences, and almost none of it
is structured around how an actual engagement flows. When you're
in the middle of something, you don't need background reading.
You need the exact command that works right now, and you need
to know what to try next if it doesn't.

That's what I built OpsecAtlas to fix.

What it actually is

It's a free, open-source penetration testing reference that
covers the full attack lifecycle. Not a wiki. Not a curated
link list. Structured methodologies built around how engagements
actually progress, with every command ready to copy and run.

The coverage spans eight areas: the full PT lifecycle from
reconnaissance to reporting, a universal methodology for any
target type, network penetration testing, web application
attacks, Active Directory compromise from initial enumeration
through DCSync and Golden Ticket, Linux privilege escalation,
Windows PrivEsc and post-exploitation, and an OWASP Top 10
attack reference.

Each section follows the same logic. A decision tree first,
so you know exactly where to start and what to try in what
order. Then the techniques with exact commands, organized by
how they would actually come up during work. Nothing else.

Why the structure matters

Most references are built to be comprehensive. This one is
built to be usable under pressure.

The Active Directory section opens with a single decision tree
that answers the question every tester asks at the start: you
have low-priv creds, now what? It walks through enumeration,
Kerberoasting, AS-REP Roasting, ACL misconfigurations, password
spraying, lateral movement paths, and DCSync in the order you
would actually work through them. Copy-ready commands at each
step. No theory upfront, no detours.

The Linux privesc section starts with an escalation priority
table so you immediately know what to check first, ordered
by how often these actually work in practice. SUID, sudo
misconfiguration, cron jobs, capabilities, writable passwd,
kernel version, all with real examples. Not textbook coverage
of every possible edge case - the things you actually hit.

The Windows section opens with the first commands you run
the moment you land a shell. Situational awareness, privilege
checking, token inspection, then the attack chains from there.

There is also a reference hub that organizes the most important
external platforms by category: HackTricks, GTFOBins, LOLBAS,
WADComs, PayloadsAllTheThings, practice labs, exploit databases.
Those external sites still need internet obviously, but having
them organized by use case instead of scattered across browser
bookmarks is genuinely useful during active work.

The offline part

It runs as a PWA. Install it once from any browser and the
entire structured content is cached locally after that. The
methodologies, decision trees, commands, and workflow notes
all work without a network connection. That matters during
exam conditions, restricted lab environments, or any situation
where you cannot freely browse.

The linked external platforms still need internet. But your
own reference stays accessible regardless.