Phase 3 — Operator Surface (Concept → Capability)

You have a substrate model from Phase 1 and you can read offense and defense as pairs from Phase 2. Phase 3 is where that knowledge becomes operator capability: workflows you can actually execute against authorized targets, not just reason about.

Phase 3 is the only phase organized around a workflow, not a body of theory. The 4 branches form a single end-to-end loop:

Mapping the surface → gathering public evidence → walking through a foothold → executing repeatable procedures.

The 41 atomic notes in those branches collapse into ~12 first-pass notes that give you the working operator loop, then ~12 more that turn the loop into a polished engagement workflow.

The Phase 3 loop

  ┌─────────────────────────────────────────────────────────┐
  │                                                         │
  ▼                                                         │
ATTACK SURFACE MAPPING  →  OSINT  →  LINUX PRIVESC  →  PLAYBOOKS  →
  (what is exposed)      (public      (post-foothold      (repeatable
                          evidence)    boundary failure)   procedures)

You can enter the loop at any node depending on engagement context. The first-pass reading order below walks the loop in its natural sequence.

First-pass reading order (12 notes, ~2 weeks)

Attack Surface Mapping — 3 notes

Attack Surface Mapping — the core mental model: surface as the intersection of reachability, discoverability, and drift.
External Attack Surface — what is exposed to the public internet, with a discipline for separating intended from actual.
Exposed Service Triage — how to rank a hundred discovered services in priority order so you spend time on the right ones.

OSINT — 3 notes

OSINT — public-source evidence gathering, with a discipline for converting clues into validated facts rather than guesses.
OSINT Triage — which findings deserve investigation now, which are noise, and how to record the decision.
Google Dorking — the most concrete OSINT technique; teaches you what search operators actually do and where indexed data leaks.

Linux Privilege Escalation — 3 notes

Linux Privilege Escalation — the core mental model: local privilege boundaries, what fails after a foothold, how enumeration becomes ranked hypotheses.
Linux Enumeration — the breadth-first inventory that turns "I have a shell" into "I have a ranked list of paths upward".
Sudo Misconfigurations — by frequency, the #1 real-world Linux privesc path. Teaches the "trust delegation" pattern that recurs everywhere else.

Security Playbooks — 3 notes

Security Playbooks Index — read this first to see the playbook template and the catalogue.
Exploit IDOR — read one offense playbook end-to-end to see how concept becomes procedure. IDOR is the right one because it is the most common real-world web finding.
Run External Recon Scan Pipeline — the engagement-level recon playbook that ties Phases 1–3 together. Read this last because it depends on everything else.

Stop here on first pass. With these 12 notes you have a working operator loop and can execute a basic external recon engagement on an authorized target end-to-end.

Extended reading (12 more notes, depth on top of first-pass)

Read as need-driven, by branch, when your engagement actually requires it.

Attack Surface Mapping depth — 4 notes

Endpoint Discovery — when web/API surface dominates the engagement.
Admin Interface Discovery — the high-impact subset of endpoint discovery.
Subdomain Takeover — the canonical dangling-DNS bug class; high-impact and findable.
Third-Party Exposure — supply-chain attack surface; often the largest unmanaged component of an organization's posture.

OSINT depth — 3 notes

Company OSINT — organization-level mapping when the engagement is corporate-scoped.
Breach and Leak Intelligence — credential exposure as a primary recon source.
OSINT Reporting — turning evidence into structured findings that survive review.

Linux Privesc depth — 3 notes

SUID/SGID Misconfigurations — the second most common privesc class in real-world findings.
Linux Capabilities — the modern replacement for SUID and a common source of subtle privesc.
LinPEAS Workflow — the practical tooling layer that converts theory into a 60-second enumeration pass.

Playbooks depth — 2 notes

Exploit SQL Injection — the canonical injection playbook; transfers to command injection, NoSQL injection, etc.
Investigate SSRF — the most-impactful cloud-era web finding; the playbook teaches the metadata-endpoint chain that turns SSRF into compromise.

How Phase 3 connects back to Phase 0 / 1 / 2

Phase 3 is where the earlier phases pay off. Use them as lenses:

From Phase 0: Every operator finding gets named with its CIA property before reporting. Every workflow gets a threat-model pass before execution. Every offensive move is paired with its defensive counterpart before claiming it succeeded.
From Phase 1: Networking and Web Security substrate is what makes attack-surface mapping intelligible; Cryptography correctness is what makes "the JWT validates" claims credible.
From Phase 2: Every Phase 3 playbook has a Phase 2 detection counterpart. The senior operator runs the offensive playbook and knows what the SOC saw.

What to skip on first pass

Phase 3 has 41 notes; the first-pass 12 + extended 12 = 24. The other 17 are depth:

OSINT specialty notes (email-and-phone-osint, image-and-location-osint, social-media-osint, historical-internet-artifacts) — workflow-specific; return when the engagement type demands them.
Attack Surface Mapping specialty (deprecated-api-versions, exposed-storage, internal-attack-surface) — return when the engagement has those surface types.
Linux Privesc specialty (path-hijacking, kernel-exploit-triage, cron-and-timer-abuse) — return after sudo and SUID basics are reflex.
Niche playbooks (test-client-ip-spoofing, test-cors-behavior, test-path-traversal, inspect-session-handling, etc.) — return when the engagement scope includes them.

This list is not a value judgment. It is operator-prioritization for someone who wants to execute one authorized engagement end-to-end before specializing.

What "first-pass complete" means in Phase 3

You have completed first-pass Phase 3 when you can:

Walk the loop from attack surface mapping → OSINT → privesc-after-foothold → playbook execution against an authorized target without an external recipe.
Triage findings — given 100 services, name which 10 deserve immediate attention and why.
Execute one playbook from memory — pick one of exploit-idor, run-scan-pipeline, or exploit-sqli and run it from start to "validated finding or negative result" without reading the page.
Write the report — turn the run into a finding with reproduction steps, CIA property, blast radius, and remediation, in under a page.

These four steps are the unit of operator capability. Phase 4 specialization is built on top of this unit.

What's next

After Phase 3 first-pass, you choose Phase 4 specialty tracks based on your job:

API Security if you build or test APIs.
Cloud Security if your systems run in cloud.
DevSecOps if you own a build/release pipeline.
Wireless Security if you work with Wi-Fi networks.

A future phase-4-specialty.md will surface the right starting notes for each. Until then, the API Security, Cloud Security, DevSecOps, and Wireless Security indexes carry their own ordered reading lists.

The Always-on parallel discipline of Privacy, Anonymity & OPSEC becomes professionally relevant in Phase 3 — every engagement leaves operator-side artifacts that need OPSEC discipline.

Start Here — persona-driven triage page.
Phase 1 — Substrate — previous phase entry.
Phase 2 — Offense / Defense — previous phase entry.
Phase 0 — Foundations — the mental models Phase 3 operationalizes.
Attack Surface Mapping Index — full branch.
OSINT Index — full branch.
Linux Privilege Escalation Index — full branch.
Security Playbooks Index — full branch.
Must-Know 30 — cross-branch must-know list.
Cybersecurity Index — full atlas roadmap.