“The adversary suffers when they cannot hide, and every move is seen.”
Organizations must evolve into a converged data and signals approach to illuminate complex attack patterns and reduce adversary dwell time.
Module 03 covers Function THREE, Converged Detection Engineering. Below: where the learner begins (what F02 and earlier produced), the work this module performs, and where the learner ends.
A complete F01 structural decomposition and an F02 mission-specific threat catalogue. No attack paths enumerated yet; you know what threats apply but not how an adversary would traverse the platform to realise them.
A complete map of attack paths (AN-ATT) drawn from the F02 threat catalogue, each spanning the chain of structural elements an adversary would touch from initial access to objective. Ready for F04 (Module 04) to enumerate the data and signal sources needed to detect each path step.
Module 03 hands-on objectives. Each row maps a LABS component to its KSAT type, (L)EARN to Knowledge, (A)PPLY to Skill, (B)UILD to Ability, (S)IMULATE to Task, so the exam at the end of the module assesses the same competencies the labs build.
| LABS Component | KSAT Type | Statement |
|---|---|---|
| (L)EARN | Knowledge | Knowledge of the AN-ATT taxonomy element, its TOE chain across multiple structural elements, and the distinction between AN-ATT (modeled or validated path) and AN-IOA (observed indicator of attack). |
| (L)EARN | Knowledge | Knowledge of how an AN-THR drives one or more AN-ATT enumerations, and how reachability across structural elements constrains valid path geometry. |
| (A)PPLY | Skill | Skill in mapping attack paths from each AN-THR through the platform's structural elements, identifying entry point, traversal sequence, and objective. |
| (A)PPLY | Skill | Skill in validating reachability between TOE steps and using red-team exercise findings to raise path confidence from modeled to validated. |
| (B)UILD | Ability | Ability to design multi-element attack paths and capture them as AN-ATT enumerated elements with full TOE chains. |
| (S)IMULATE | Task | Enumerate the AN-ATT map for a sample LEO platform covering every AN-THR from F02, ready for F04 to drive detection engineering. |
Function Three elements two distinct analytic statements: AN-ATT, what the platform exposes, conditionally, and AN-IOA, what an adversary is doing against the platform right now. Same TOE field, two different tenses; together they bracket the past (AN-IOC) and the future (AN-THR).
Modeled or known attack paths through the platform, with TOEs spanning every structural layer the path traverses.
In-progress attack indicators tied to live telemetry, with TOEs naming the structural entries currently exposed.
The framework elements both what could happen (AN-ATT) and what is happening (AN-IOA) without collapsing them.
Function Three elements two distinct analytic statements: AN-ATT, what the platform exposes, conditionally, and AN-IOA, what an adversary is doing against the platform right now. Same TOE field, two different tenses; together they bracket the past (AN-IOC) and the future (AN-THR).
Module 1 structural CONOPS + Module 2 AN-THR elements.
AN-ATT entries (modeled exposure paths) and AN-IOA entries (verifiable in-progress attack indicators), both attached via TOE.
Function THREE attaches its work to the structural decomposition produced upstream, the AN-THR threat catalogue and the CONOPS structural model. Every AN-ATT you enumerate in this module attaches via TOE to one or more structural elements. Quick recap of the four structural layers and the analytic overlay.
PCEPrimary Capability EnvironmentWhere the platform physically operates. Five environments: Terrestrial, Aquatic, Aerial, Orbital, Deep Space.
SEGSegmentSelf-contained piece with a specific operational role. Ten segments including Launch, Link, Ground, User, Space, Deep Space.
SVCServiceThe functional plane, how a segment controls things or moves data. Three services: Control Plane, Data Plane, Hybrid.
ASTAssetConcrete things that make a service work. Six asset classes: Hardware, Firmware, Software, Data, Signal, Hybrid.
ANAnalytic (this module's home)A separate overlay layer. Carries what defenders observe and build. Six categories, AN-ATT is the one this module produces.
Same AN-ATT two ways. The taxonomy element is the abstract category, written hyphenated as AN-ATT. The enumerated element is one specific instance on your platform, written with all five fields plus a description, e.g. AN: ATT: Attack Path: 00. Use the hyphenated form when you mean “any Attack Path”; use the full form when you mean “this exact Attack Path on our platform.”
The category. Written LAYER-ELEMENT (hyphenated). Use it in prose to refer to any Attack Path.
AN-ATT
“Every AN-ATT entry must have a documented source.”, talking about attack paths in general.
AN-ATT
“Our AN-ATT catalogue is reviewed quarterly against intel updates.”
One specific instance. Written LAYER: ELEMENT: LABEL: ORDINAL with description and TOE.
AN: ATT: Attack Path: 00
“AN: ATT: 00, Ground-to-Space Command Injection Path”
AN: ATT: Attack Path: 01
A second specific attack path on the same platform, same taxonomy code, different ordinal, different description and TOE.
Each AN-ATT enumerated element points back at the platform via the TOE (Target of Exploitation) field. TOE names the chain of structural elements the adversary touches. Without TOE, attack paths are free-floating noise, the structural anchor is what makes the attack path actionable, queryable, and shareable.
Each AN-ATT element carries a TOE field that lists the structural element(s) the attack path chain of structural elements the adversary touches. Format: TOE: structural element references.
AN: ATT: 00, Ground-to-Space Command Injection Path: TOE: AST: SW: Software: 03 (entry), SVC: CP: Control Plane: 00 (traversal), SEG: SP: Space: 00 (objective). Every structural anchor is a real enumerated element on the platform, never a hypothetical, never a sample.
An AN-ATT with no TOE attachment is free-floating noise. Every entry must point at one or more real structural elements on your platform. This is the discipline that keeps the analytic catalogue queryable, correlatable, and shareable.
Per-element enumeration procedure. The walk is the same for every AN-ATT; only the inputs and the structural anchors change. Sources: the AN-THR threat catalogue and the CONOPS structural model.
01020304050601 · PICK AN AN-THR
Start from the F02 catalogue. Each AN-THR may yield one or more attack paths.
02 · IDENTIFY ENTRY POINT
Where does the adversary first touch the platform to realize this threat? Pick the structural element (PCE / SEG / SVC / AST) of initial access.
03 · MAP THE CHAIN
List the structural elements the adversary must traverse from initial access to objective. The chain becomes the TOE list.
04 · VALIDATE AGAINST ARCHITECTURE
Each step must be reachable from the previous. If a TOE element is unreachable, the path is invalid, revise.
05 · ENUMERATE AN-ATT
Create the AN-ATT enumerated element with the full TOE chain and a description.
06 · COVERAGE ANALYSIS
For each AN-THR, count the AN-ATT paths. Threats with no path: untraversable on this platform, or path missing, investigate.
A real-world attack path for an orbital constellation, end-to-end, the enumerated element, the TOE attachment, the sourcing, and how it hands off to the next function.
A validated path from AN: THR: 00 (the LEO-targeting APT). Adversary gains foothold on a ground software asset, traverses to the control plane, sends commands to the space segment.
AN: ATT: Attack Path: 00
TOE: AST: SW: Software: 03 (entry), SVC: CP: Control Plane: 00 (traversal), SEG: SP: Space: 00 (objective)Driven by AN: THR: 00; path validated in red-team exercise 2026-Q1
Real-world validated only. No hypotheticals. The sourcing trail makes the entry auditable and lets analysts revisit when intel evolves.
F04 enumerates data and signal sources for each step, then writes RootA.io detection signatures.
F04 (Incident Response Preparation), enumerates AN-DET data sources and detection signatures for each AN-ATT.
Before publishing an AN-ATT to your TIP or sharing through Space ISAC, verify:
An enumerated AN-ATT carries five core fields plus the TOE (Target of Exploitation) attachment that makes it actionable. Cycle through each below to see what the field holds, what a real value looks like, and where learners typically slip.
LAYERField 1 of 5Fixed for every analytic-layer entry.
AN
ELEMENTField 2 of 5Two-letter taxonomy code identifying the path sub-category.
ATT
LABELField 3 of 5Plain-English name for the ATT code.
Attack Path
ORDINALField 4 of 5Two-digit serial; first attack path mapped is 00.
00
DESCRIPTIONField 5 of 5Free-text scoping with the realistic narrative the path describes.
"Phishing → ground software → control-plane traversal → command issuance to space segment, validated in red-team exercise 2026-Q1."
TOETOE attachment (AN-specific)TOE on AN-ATT spans multiple structural elements, the chain the adversary touches from initial access to objective.
TOE: AST: SW: Software: 03, SVC: CP: Control Plane: 00, SEG: SP: Space: 00
Four worked AN-ATT enumerations spanning different scenarios on the same LEO platform, nation-state, supply chain, RF, insider. Each one is real-world validated, structurally anchored via TOE, and traceable to its source.
Initial access via phishing on ground-station SW; lateral movement to control-plane service; command issuance to space segment.
AN: ATT: Attack Path: 00
TOE: AST: SW: Software: 03, SVC: CP: Control Plane: 00, SEG: SP: Space: 00
Sourcing: Driven by AN: THR: 00 (LEO-targeting APT); validated in red-team 2026-Q1
Compromised SW build → firmware payload → onboard bus hardware. Not yet observed; modeled from architecture review.
AN: ATT: Attack Path: 01
TOE: AST: SW: Software: 03 → AST: FW: Firmware: 00 → AST: HW: Hardware: 00
Sourcing: Driven by AN: THR: 01 (supply-chain compromise); modeled against architecture
Adversary jams uplink signal → link layer failure → data-plane mission outage. Cross-segment chain.
AN: ATT: Attack Path: 02
TOE: AST: SI: Signal: 00, SEG: LI: Link: 00, SVC: DP: Data Plane: 00
Sourcing: Driven by AN: THR: 02 (RF jamming campaign); observed during contested ops
Privileged insider physically tampers with ground hardware; uses elevated access to inject control-plane commands.
AN: ATT: Attack Path: 03
TOE: SEG: GR: Ground: 00, AST: HW: Hardware: 03, SVC: CP: Control Plane: 00
Sourcing: Driven by AN: THR: 03 (insider profile); validated in post-incident debrief
AN-ATT · the conditional statementKnown or modeled attack path for a converged space system, a sequence of adversary actions that could traverse the platform from entry to impact.
| LAYER | ELEMENT | LABEL | DESCRIPTION |
|---|---|---|---|
AN | ATT | Attack Path | Known or modeled attack path for a converged space system, a sequence of adversary actions that could traverse the platform from entry to impact. |
TOETarget of Exploitation, lists every structural entry the path traverses end-to-end. AN-ATT TOEs are typically longer than AN-IOC or AN-IOA TOEs because the path itself spans multiple layers.
Sourcing: Adversary-behavior catalogs (ATT&CK, ATLAS, SPARTA), red-team reports, internal architecture reviews, and peer-shared attack-path documentation.
By giving Attack Paths their own analytic category distinct from indicators, METEORSTORM lets the same defender reason about exposures that have been modeled but never exercised, exposures exercised by red teams, and exposures exploited in the wild, all in the same enumeration. The structural answer to “what could happen here?” is enumerated as cleanly as the answer to “what is happening here?”
LAYER = AN (fixed).
Adversary-behavior catalog (ATT&CK, ATLAS, SPARTA), red-team report, architecture review, or peer-shared documentation.
Identifies this as an Attack-Path entry within the Analytic Layer.
Two-digit, AN-ATT-00, AN-ATT-01, …
List every structural entry the path traverses, end-to-end. AN-ATT TOEs are typically longer because the path spans multiple layers.
Describe the path in enough detail for another analyst to evaluate it against their own platform. Cite the source framework or report.
AN-IOA · the present-tense statementVerifiable indication that the platform has been targeted, adversary activity in progress against the platform, observable right now.
| LAYER | ELEMENT | LABEL | DESCRIPTION |
|---|---|---|---|
AN | IOA | Indicator of Attack | Verifiable indication that the platform has been targeted, adversary activity in progress against the platform, observable right now. |
TOETarget of Exploitation, lists the structural entries currently exposed to the in-progress attack, typically the SVC and SEG involved, plus any AST in scope.
Sourcing: Operator telemetry, confirmed alerts, or incident reports from trusted partners. Modeled behavior without a current observation belongs in AN-ATT, not AN-IOA.
Most defensive frameworks collapse the present and the past into a single “detection” or “alert” concept. METEORSTORM keeps them separate so the analyst can answer two distinct questions independently: did something already happen, and is something happening right now? Two questions, two analytic categories, two decision paths.
LAYER = AN (fixed).
Live telemetry, confirmed alert, or trusted-partner incident report. If not currently observed, the entry belongs in AN-ATT.
Identifies this as an Indicator-of-Attack entry within the Analytic Layer.
Two-digit, AN-IOA-00, AN-IOA-01, …
List the structural entries currently exposed to the in-progress attack: typically the SVC and SEG involved, plus any AST in scope.
Describe the in-progress activity (pattern, cadence, source where applicable) and cite the observation source.
Module 04 enumerates the data and signal sources needed to detect each AN-ATT path, then writes detection signatures in RootA.io format.
A multiple-choice exam aligned with Module 03 KSAT areas. Drawn at random from a question bank covering Function THREE's taxonomy element (AN-ATT), its TARGET attachment (TOE), and the production flow into the next function. Exam scaffolding wired in next iteration.
20 questions, drawn at random from a 20-question bank, aligned with Module 03 KSAT areas: Knowledge, Skills, Abilities, and Tasks. Question and answer order are randomized each session.