Book Blueprint

This blueprint records the editorial intent of Tools: A Reference to the Templates, Software and Methods of The Transformed Architecture for European Enterprises and Public Authorities, a companion to The Transformed Architecture. The authoritative design is DISPOSITION.md; the chapter briefs below are generated from scripts/outline_data.py and data/tool_inventory.jsonl by scripts/generate_outlines.py.

Part I: The Toolbox

Chapter 1: Introduction: A Toolbox for the Series

File: 01_introduction.md
Status: draft
Purpose: Define what a tool is in the series, state the book's rules and give each reader a path through the catalogue.
Draws on: Book 1, Book 2, Book 3, Book 4, Book 5, Book 6, Book 7, Book 8, Book 9, Book 10
Inventory: no inventoried entries
Questions: What counts as a tool, and what does not? How is the book organised and how should each kind of reader use it?
Planned sections: Four kinds of tool; How the book was compiled; How to read an entry; Nordvik Region and Hansa Mutual; Summary
Worked example: A reader finds, in ten minutes, the tools that her next incident review needs.
Figure: The four kinds of tool and the eight activities of the toolbox

Chapter 2: The Anatomy of a Tool: Kinds, Identifiers and Lifecycle

File: 02_the_anatomy_of_a_tool.md
Status: draft
Purpose: Describe the anatomy that every entry shares: purpose, inputs, outputs, identifier family, owner, evidence age and review date.
Draws on: Book 1, Book 2, Book 3, Book 6, Book 10
Inventory: no inventoried entries
Questions: What do the identifier families (SVC-, DLG-, MND-, RoE- and the rest) have in common? What is the lifecycle of a tool from first use to retirement?
Planned sections: The common anatomy; Identifier families and how they link; Evidence age on every tool; Tools in version control; Summary
Worked example: Tracing one service record to the delegation profile, the observer manifest and the restore test that refer to it.
Figure: How the identifier families link across the books

Chapter 3: Choosing, Adopting and Retiring Tools

File: 03_choosing_adopting_and_retiring_tools.md
Status: draft
Purpose: Give a procedure for choosing, adopting and retiring tools, with the evidence each step needs.
Draws on: Book 1, Book 2, Book 3, Book 10
Inventory: 24 artefacts, 22 methods, 9 third-party tools and standards
Questions: How does an organisation choose between a template, a product and a method? What proves that a tool is adopted and still works?
Planned sections: Selecting; Adoption and the diffusion of tools; Maintenance and review; Retirement and exit; Anti-patterns; Summary
Worked example: Adoption roadmap for a policy engine at Nordvik Region.
Figure: Tool lifecycle with decision points

Part II: Architecture and Delivery

Chapter 4: Architecture and Decision Tools

File: 04_architecture_and_decision_tools.md
Status: draft
Purpose: Collect the tools for modelling architecture, recording decisions and documenting as code.
Draws on: Book 1, Book 2, Book 3, Book 6
Inventory: 22 artefacts, 12 methods, 7 own tools, 29 third-party tools and standards
Questions: Which tools describe architecture so that it can be versioned, reviewed and tested? Which records carry the reasons for a decision?
Planned sections: Architecture modelling; Architecture and decision records; Documentation, requirements and knowledge as code; Across the usage modes; Summary
Worked example: An ADR and a C4 model reviewed in the same pull request.
Figure: From model to decision record

Chapter 5: Policy, Pipeline and Evidence Tools

File: 05_policy_pipeline_and_evidence_tools.md
Status: draft
Purpose: Collect the tools that express policy as code, run pipelines and produce evidence.
Draws on: Book 1, Book 2, Book 3, Book 4, Book 6, Book 10
Inventory: 84 artefacts, 28 methods, 3 own tools, 54 third-party tools and standards
Questions: Which tools turn a rule into a check that runs on every change? Which tools produce the evidence an auditor can verify?
Planned sections: Policy engines and rules; Pipelines and testing; Supply-chain tools: provenance, signing, bills of materials; Evidence manifests; Across the usage modes; Summary
Worked example: A pipeline gate that rejects an unsigned model artefact.
Figure: Policy, pipeline and evidence chain

Chapter 6: Platform and Infrastructure Tools

File: 06_platform_and_infrastructure_tools.md
Status: draft
Purpose: Collect the platform and infrastructure tools: infrastructure as code, containers, MLOps, model serving, retrieval and gateways.
Draws on: Book 1, Book 3, Book 10
Inventory: 33 artefacts, 3 methods, 33 third-party tools and standards
Questions: Which tools build and run the AI platform? Which choices are open standards and which are products?
Planned sections: Infrastructure as code; Container and service-mesh platforms; MLOps and model serving; Retrieval and data tools; Gateways; Summary
Worked example: Platform baseline for a regional AI platform.
Figure: Layers of the AI platform and their tools

Part III: Governing AI

Chapter 7: Inventory, Classification and Risk Tools

File: 07_inventory_classification_and_risk_tools.md
Status: draft
Purpose: Collect the tools that inventory AI use, classify it and manage its risk.
Draws on: Book 2, Book 3, Book 4, Book 7, Book 10
Inventory: 57 artefacts, 22 methods, 3 third-party tools and standards
Questions: Which registers and records does a governed organisation keep? How is an exception or an impact assessment recorded?
Planned sections: Inventory and intake; Classification; Risk and impact assessment; AI Decision Records; Summary
Worked example: An intake record that classifies a use case as high-risk.
Figure: From intake to decision record

Chapter 8: Obligation, Assurance and Maturity Tools

File: 08_obligation_assurance_and_maturity_tools.md
Status: draft
Purpose: Collect the tools that map obligations, provide assurance and assess maturity.
Draws on: Book 1, Book 2, Book 3, Book 8, Book 10
Inventory: 65 artefacts, 29 methods, 1 own tools, 37 third-party tools and standards
Questions: Which tools map an obligation to a control and a piece of evidence? Which tools measure maturity and what do they omit?
Planned sections: Obligation mapping; Assurance and audit evidence; Maturity models and radar tools; Reporting and transparency; Summary
Worked example: A control mapping row linking one control to three frameworks.
Figure: Assure once, comply many

Chapter 9: Supplier, Contract and Sovereignty Tools

File: 09_supplier_contract_and_sovereignty_tools.md
Status: draft
Purpose: Collect the tools for procurement, contracts, sovereignty and exit.
Draws on: Book 2, Book 3, Book 4, Book 10
Inventory: 21 artefacts, 8 methods, 5 third-party tools and standards
Questions: Which clauses, questionnaires and registers does procurement of AI need? How is exit tested?
Planned sections: Procurement; Contract clauses; Sovereignty assessment; Exit and concentration; Summary
Worked example: A sovereignty assessment of a model provider.
Figure: Supplier lifecycle with exit test

Part IV: Securing and Testing

Chapter 10: Threat Modelling and Adversary Frameworks

File: 10_threat_modelling_and_adversary_frameworks.md
Status: draft
Purpose: Collect the frameworks and methods for modelling threats to AI systems and to the organisation.
Draws on: Book 1, Book 3, Book 4, Book 5, Book 6, Book 8, Book 9
Inventory: 4 artefacts, 5 methods, 44 third-party tools and standards
Questions: Which frameworks describe adversary behaviour at the right level? Which methods turn a threat model into tests?
Planned sections: Adversary frameworks; Threat modelling methods; Lists of AI security risks; Threat registers; Summary
Worked example: A threat model for a retrieval-augmented assistant.
Figure: From framework to threat model to test

Chapter 11: AI Red Teaming and Emulation Tools

File: 11_ai_red_teaming_and_emulation_tools.md
Status: draft
Purpose: Collect the tools for AI red teaming and adversary emulation, defensive and authorised only.
Draws on: Book 3, Book 4, Book 6, Book 8, Book 9
Inventory: 8 artefacts, 10 methods, 7 third-party tools and standards
Questions: What must a rules-of-engagement manifest contain? Which tools evaluate AI systems under authorised test?
Planned sections: Rules of engagement; Emulation plans; Evaluation and red-team harnesses; Reporting findings; Summary
Worked example: A rules-of-engagement manifest for a contracted test.
Figure: Authorised test from scope to finding

Chapter 12: Detection, Investigation and Intelligence Tools

File: 12_detection_investigation_and_intelligence_tools.md
Status: draft
Purpose: Collect the tools for detection engineering, investigation, hunting and intelligence.
Draws on: Book 4, Book 6, Book 8, Book 9
Inventory: 38 artefacts, 21 methods, 39 third-party tools and standards
Questions: Which formats and records carry a detection from idea to retirement? How is an AI-assisted investigation recorded?
Planned sections: Detection records and rules; Case and playbook records; Hunt records; Intelligence and vulnerability data; Across the usage modes; Summary
Worked example: A detection record with test cases and a retirement date.
Figure: Detection lifecycle

Chapter 13: Response, Recovery and Resilience Tools

File: 13_response_recovery_and_resilience_tools.md
Status: draft
Purpose: Collect the tools for incident response, recovery and resilience.
Draws on: Book 3, Book 4, Book 5, Book 8, Book 9, Book 10
Inventory: 29 artefacts, 25 methods, 10 third-party tools and standards
Questions: Which records carry an incident from detection to review? How is recovery proved and how old is the proof?
Planned sections: Incident records and reporting clocks; Recovery plans and restore tests; Post-incident review; Resilience registers; Summary
Worked example: A restore test whose result is two years old.
Figure: Incident to review to retest

Chapter 14: The Verification Loop: Purple, Chaos and Coverage Tools

File: 14_the_verification_loop.md
Status: draft
Purpose: Collect the tools of the verification loop: purple sessions, coverage records, regression cases and security chaos engineering.
Draws on: Book 1, Book 4, Book 9
Inventory: 21 artefacts, 21 methods, 9 third-party tools and standards
Questions: What makes a coverage claim evidence rather than assertion? How is independence between tester and tested kept?
Planned sections: Purple sessions; Coverage and evidence age; Regression cases; Security chaos engineering; Summary
Worked example: A coverage record with evidence age.
Figure: The purple loop and its five stages

Part V: Operating

Chapter 15: Observability and Service Objective Tools

File: 15_observability_and_service_objective_tools.md
Status: draft
Purpose: Collect the tools for observability and service objectives.
Draws on: Book 2, Book 3, Book 10
Inventory: 12 artefacts, 16 methods, 1 own tools, 14 third-party tools and standards
Questions: How are objectives, alerts and error budgets recorded? How does AI correlation stay within a noise budget?
Planned sections: Telemetry and standards; Service level objective records; Alerting and burn rates; Correlation policy and noise budget; Measurement; Summary
Worked example: A service level objective record with burn-rate alerts.
Figure: From telemetry to page

Chapter 16: Change, Restore and Runbook Tools

File: 16_change_restore_and_runbook_tools.md
Status: draft
Purpose: Collect the tools for change, restoration and runbooks.
Draws on: Book 1, Book 3, Book 10
Inventory: 19 artefacts, 21 methods, 12 third-party tools and standards
Questions: How is a change classified, frozen and rolled back? How is a runbook made executable and reviewable?
Planned sections: Change classes and records; Progressive delivery and GitOps; Runbooks and remediation; Freezes and windows; Summary
Worked example: A runbook manifest executed at Level 3.
Figure: Change to restore with stop path

Chapter 17: Capacity, Cost and Energy Tools

File: 17_capacity_cost_and_energy_tools.md
Status: draft
Purpose: Collect the tools for capacity, cost and energy.
Draws on: Book 1, Book 2, Book 3, Book 10
Inventory: 14 artefacts, 5 methods, 12 third-party tools and standards
Questions: Which records steer the cost of AI services? How is energy accounted for?
Planned sections: Cost management and FinOps; Capacity records; Token and GPU economics; Energy and carbon; Summary
Worked example: A consumption forecast with a budget alert.
Figure: Cost steering loop

Part VI: Agents and Autonomy

Chapter 18: Mandate, Delegation and Autonomy Tools

File: 18_mandate_delegation_and_autonomy_tools.md
Status: draft
Purpose: Collect the tools that bound what an agent may do: mandates, delegation profiles, autonomy levels and budgets.
Draws on: Book 2, Book 3, Book 6, Book 7, Book 8, Book 10
Inventory: 39 artefacts, 24 methods, 1 own tools, 12 third-party tools and standards
Questions: What does a mandate manifest contain? How is an autonomy level raised or lowered?
Planned sections: Mandate manifests; Delegation profiles; Autonomy budgets; Promotion and demotion; Across the degrees of delegation; Summary
Worked example: A delegation profile for an operations agent.
Figure: Mandate, delegation and budget

Chapter 19: Observer, Stop Path and Evaluation Tools

File: 19_observer_stop_path_and_evaluation_tools.md
Status: draft
Purpose: Collect the tools that verify agents independently: observers, stop paths, seeded faults and evaluations.
Draws on: Book 4, Book 6, Book 9, Book 10
Inventory: 23 artefacts, 25 methods, 2 third-party tools and standards
Questions: What must be independent of what it verifies? Where does the stop path run?
Planned sections: Observer manifests; Seeded faults; Stop paths and drills; Agent evaluation; Summary
Worked example: An observer manifest and a seeded fault.
Figure: Observer principle

Chapter 20: Defending the Agent: Identity, Credential and Protocol Tools

File: 20_defending_the_agent.md
Status: draft
Purpose: Collect the tools that secure the agent itself: identity, credentials, protocols, containment and tier-0 controls.
Draws on: Book 3, Book 4, Book 6, Book 7, Book 8, Book 10
Inventory: 29 artefacts, 9 methods, 20 third-party tools and standards
Questions: Which identity and protocol standards do agents rely on? Which controls make the operator's AI tier 0?
Planned sections: Agent identity and workload identity; Credentials and secrets; Agent protocols; Containment and permission manifests; Summary
Worked example: A permission manifest for a coding agent.
Figure: Agent identity and containment

Part VII: People, Organisation and Society

Chapter 21: Role, Competence and Career Tools

File: 21_role_competence_and_career_tools.md
Status: draft
Purpose: Collect the tools for roles, competence and careers.
Draws on: Book 1, Book 2, Book 6, Book 8, Book 9, Book 10
Inventory: 31 artefacts, 14 methods, 7 third-party tools and standards
Questions: How is a role charter written and reviewed? How is competence recorded without scoring people?
Planned sections: Role charters; Competence records; Learning paths and training; Keeping the craft; Summary
Worked example: A role charter and a competence record for a conductor role.
Figure: Role, competence and career

Chapter 22: Tools for Managing People with AI: Decision, Contestation and Care

File: 22_tools_for_managing_people_with_ai.md
Status: draft
Purpose: Collect the tools for the governance of AI that manages people: decision, delegation, contestation and care.
Draws on: Book 2, Book 7
Inventory: 22 artefacts, 11 methods
Questions: Which records give a named human decider? How is a decision contested?
Planned sections: Management decision records; Contestation and appeal; Impact tests with representatives; Care and wellbeing; Summary
Worked example: A contestation record and an appeal path.
Figure: Decision, contestation and appeal

Chapter 23: Tools for Weighing Evidence: Threat Registers, Debate and Foresight

File: 23_tools_for_weighing_evidence.md
Status: draft
Purpose: Collect the tools for weighing evidence about threats and the future: evidence ladder, threat register, constructive debate and the three outlook labels.
Draws on: Book 2, Book 5, Book 8, Book 9, Book 10
Inventory: 20 artefacts, 24 methods, 3 third-party tools and standards
Questions: How is a claim graded? How is a debate staged so that it ends in an action?
Planned sections: The evidence ladder; Threat registers; Constructive debate in four steps; Observable, expected, speculation; Summary
Worked example: A threat register entry graded on the evidence ladder.
Figure: Evidence ladder and debate

Part VIII: Experience and Conclusions

Chapter 24: Building a Toolbox: Nordvik Region and Hansa Mutual

File: 24_building_a_toolbox.md
Status: draft
Purpose: Show how Nordvik Region and Hansa Mutual build and run their toolboxes.
Draws on: Book 2, Book 3, Book 4, Book 10
Inventory: no inventoried entries
Questions: Which tools do the two organisations adopt first, and why? Where do they differ?
Planned sections: Nordvik Region; Hansa Mutual; Compared; Summary
Worked example: A year of tool adoption at the two organisations.
Figure: Adoption sequence at the two organisations

Chapter 25: Conclusions: One Toolbox for the Series

File: 25_conclusion.md
Status: draft
Purpose: Draw the book's own conclusions about the toolbox of the series.
Draws on: Book 1, Book 2, Book 3, Book 4, Book 5, Book 6, Book 7, Book 8, Book 9, Book 10
Inventory: no inventoried entries
Questions: What does the toolbox show about the series? What gaps remain?
Planned sections: What the toolbox shows; Gaps and inconsistencies found; What to do on Monday; Summary
Worked example: A gap list for the editor of each earlier volume.
Figure: Toolbox on one page