Skip to content

CFSE - Concept Flow Scenarios Explorations

Version 1.0.0 | January 2026

CFSE is a formal methodology for security epistemology: a disciplined approach to replacing intuition with structure, and claims with evidence.

CFSE is a formal methodology for security epistemology—a disciplined approach to knowing whether a system is secure, not guessing or hoping. It builds a world model of systems through concepts, interactions, and flows, expresses security properties as formal invariants with temporal logic, and validates claims through systematic exploration backed by structured evidence.

Not a scanner. Not a checklist. A formal framework for security knowledge.


Quick Start

GoalResource
Start here (purpose + model)guides/foundations/introduction.md
Understand the artifact pipelinespec/foundations/artifact-system.md
Conformance levelsspec/foundations/conformance.md
ID syntaxspec/grammar/01-id-syntax.md
Normative specspec/
Architecture mapARCHITECTURE.yml

Directory Structure

cfse-spec/
spec/ # Normative core specification (source of truth)
foundations/ # Core concepts: artifact system, conformance, extensions
grammar/ # ID syntax, references, formal logic, field types
artifacts/ # Artifact definitions with YAML schemas
primary/ # Concept, Interaction, Flow, Scenario, Exploration, Finding
supporting/ # Predicate, Invariant, Generator, Patch
semantics/ # Invariant states, verdicts, lifecycle, traceability
extensions/ # Optional extensions (opt-in; may be empty)
guides/ # Informative background and reading guides
rfcs/ # Informative RFC/proposal process (no archive required)
ARCHITECTURE.yml # Machine-readable structure manifest

The CFSE Pipeline

Concept -> Interaction -> Flow -> Scenario -> Exploration -> Finding -> Patch
C I F S E FD PATCH
  1. Concept (C-): Define system building blocks
  2. Interaction (I-): Document atomic operations
  3. Flow (F-): Map legitimate user journeys
  4. Scenario (S-): Hypothesize security violations
  5. Exploration (E-): Test hypotheses with BASE vs ATT
  6. Finding (FD-): Document confirmed vulnerabilities
  7. Patch (PATCH-): Fix, verify, and learn

Key Specifications

TopicLocation
Framework Introductionguides/foundations/introduction.md
World Modelguides/foundations/world-model.md
Artifact Systemspec/foundations/artifact-system.md
ID Syntaxspec/grammar/01-id-syntax.md
Formal Logicspec/grammar/03-formal-logic.md

Automation (Optional)

CFSE is designed to be both human-readable and tool-friendly.

  • ARCHITECTURE.yml provides a machine-readable map of the spec (useful for editors, validators, and automation).
  • Some supporting material (templates, examples, prompts) may be published separately.

Classification

  • Normative (spec/): Authoritative definitions - source of truth
  • Informative (guides/, rfcs/): Explanations, templates, and design discussion scaffolding

Positioning

CFSE is compatible with many existing practices (threat modeling, red teaming, property-based testing, and formal methods). It is not a replacement for any of them; it is the “evidence-linked structure layer” that makes security reasoning auditable and repeatable.

Depth, Not Surface

Surface approachCFSE equivalent
Scan for vulnerabilitiesBuild world model, derive hypotheses from structure
Follow a checklistExpress properties as formal invariants
Run some testsSystematic explorations with baseline/attack deltas
Write up findingsEvidence-backed findings with trace provenance

See guides/foundations/introduction.md for the full positioning.


Getting Help

  1. Start with guides/foundations/introduction.md
  2. Read spec/README.md for the spec map
  3. Consult spec/glossary.md for terminology
  4. Reference ARCHITECTURE.yml for the complete structure map

Contributing

See CONTRIBUTING.md.

About

The first formal security methodology designed for human-AI collaboration. Systematic. Evidence-based. Open source.

Topics

Resources

Code of conduct

Contributing

Security policy

Stars

3 stars

Watchers

0 watching

Forks

Contributors