{"levels":[[0,"Absent","No tool of this kind is in use."],[1,"Ad hoc","Individuals use something of the kind. Nothing is written down and nobody owns it."],[2,"Defined","A template or record from the series is adopted. A named owner exists and the artefact is in version control."],[3,"Operated","The tool is used on real work, in the pipeline or the process. Its review date is set and kept."],[4,"Verified","Someone independent of the author and the operator has tested that it works (a seeded fault, a restore test, an exercise), and the evidence is younger than its window."],[5,"Continuously assured","The checks run automatically, the evidence renews on defined triggers, and adoption, renewal and retirement are decided on measured evidence."]],"window":12,"next":{"0":"Pick the principal template for this practice in the chapter named, copy it into version control and name an owner.","1":"Write the practice down: adopt the template, name an owner and put the artefact in version control.","2":"Use it on real work: attach it to the pipeline or process and set a review date that is kept.","3":"Have someone independent of the author and the operator test it (seeded fault, restore test or exercise) and record the date and result.","4":"Automate the check, define the triggers that expire the evidence, and measure it over time.","5":"Keep the evidence current; review whether the practice is still needed."},"areas":[{"id":"A","title":"Architecture and delivery","chapters":"Chapters 4 to 6","practices":[{"id":"A1","text":"Architecture decisions are recorded as decision records in version control, with the reasons, the alternatives and a review date.","tools":"Architecture Decision Record (ADR)","chapter":4,"search":"decision record"},{"id":"A2","text":"Architecture is described as code (models, diagrams, documentation) and reviewed in the same change as the system it describes.","tools":"Architecture model as code; documentation as code","chapter":4,"search":"model"},{"id":"A3","text":"Policy is written as code, tested, and enforced as a gate in the delivery pipeline.","tools":"Policy as code; Open Policy Agent and Rego","chapter":5,"search":"policy"},{"id":"A4","text":"Each release produces an evidence manifest (provenance, signatures, bill of materials) that a third party can check.","tools":"Evidence manifest; provenance and signing","chapter":5,"search":"evidence"},{"id":"A5","text":"The platform and its AI components are defined as code from a baseline, and drift is detected.","tools":"Infrastructure as code; platform baseline","chapter":6,"search":"baseline"}]},{"id":"B","title":"Governing AI","chapters":"Chapters 7 to 9","practices":[{"id":"B1","text":"Every AI use is in an inventory with an owner, an intake record and a risk classification.","tools":"AI inventory; intake record; risk tiering","chapter":7,"search":"intake"},{"id":"B2","text":"Risks and exceptions are recorded with an owner, an expiry and a review date; high-risk uses have an impact assessment.","tools":"Risk record; exception record; impact assessment","chapter":7,"search":"risk"},{"id":"B3","text":"Decisions about AI are recorded as AI Decision Records that name who decided.","tools":"AI Decision Record (AIDR)","chapter":7,"search":"AIDR"},{"id":"B4","text":"Obligations are mapped to controls and evidence once and reused across frameworks (assure once, comply many).","tools":"Control mapping matrix","chapter":8,"search":"control"},{"id":"B5","text":"Suppliers of AI are assessed for sovereignty and contract terms, and the exit plan has been tested.","tools":"Sovereignty assessment; exit plan","chapter":9,"search":"supplier"}]},{"id":"C","title":"Securing and testing","chapters":"Chapters 10 to 14","practices":[{"id":"C1","text":"Each AI system has a threat model that uses a named framework and produces tests.","tools":"Threat model; ATT&CK and ATLAS","chapter":10,"search":"threat"},{"id":"C2","text":"Adversarial testing runs only under a signed rules-of-engagement manifest that is enforced outside the tester.","tools":"Rules-of-engagement manifest (RoE)","chapter":11,"search":"RoE"},{"id":"C3","text":"Detections are records with an owner, test cases and a review or retirement date.","tools":"Detection record (DET)","chapter":12,"search":"detection"},{"id":"C4","text":"Incidents are recorded to a standard, reporting clocks are configured, and recovery is proved by restore tests whose evidence age is known.","tools":"Incident record; restore-test record","chapter":13,"search":"restore"},{"id":"C5","text":"Coverage is claimed only with evidence (what was tested, when, with what result), and the tester is independent of the tested.","tools":"Coverage record; regression case; purple session record","chapter":14,"search":"coverage"}]},{"id":"D","title":"Operating","chapters":"Chapters 15 to 17","practices":[{"id":"D1","text":"Services have records with dependencies and blast radius, and objectives with burn-rate alerts that have been tested.","tools":"Service record (SVC); service level objective record (SLO)","chapter":15,"search":"SLO"},{"id":"D2","text":"Alert correlation and AI assistance in observation operate within a written policy and a noise budget.","tools":"Correlation policy and noise budget","chapter":15,"search":"correlation"},{"id":"D3","text":"Changes are classified, windows and freezes are enforced, and every change has a way back.","tools":"Change class; change record (CHG)","chapter":16,"search":"change"},{"id":"D4","text":"Runbooks are manifests with defined authority, a tested rollback and a stop path.","tools":"Runbook manifest (RBK)","chapter":16,"search":"runbook"},{"id":"D5","text":"Capacity, cost and energy are recorded and steered against budgets, including token and GPU cost.","tools":"Capacity and cost record (CAP); FinOps","chapter":17,"search":"cost"}]},{"id":"E","title":"Agents and autonomy","chapters":"Chapters 18 to 20","practices":[{"id":"E1","text":"Every agent has a mandate manifest that names its purpose, actions, data and the accountable person.","tools":"Mandate manifest (MND)","chapter":18,"search":"mandate"},{"id":"E2","text":"Delegation is set per action class and level, the autonomy budget is tracked, and changes are decided by a named authority.","tools":"Delegation profile (DLG); autonomy budget","chapter":18,"search":"delegation"},{"id":"E3","text":"What verifies an agent is independent of it: an observer manifest and seeded faults.","tools":"Observer manifest (OBS); seeded-fault record","chapter":19,"search":"observer"},{"id":"E4","text":"A stop path outside the agent platform exists, is held by named people and is drilled.","tools":"Stop path; stop drill","chapter":19,"search":"stop"},{"id":"E5","text":"Agents have their own identity, scoped credentials and a permission manifest, and the agent control plane is treated as Tier 0.","tools":"Workload identity; permission manifest","chapter":20,"search":"permission"}]},{"id":"F","title":"People, organisation and society","chapters":"Chapters 21 to 23","practices":[{"id":"F1","text":"Roles that use or supervise AI have charters with decision rights.","tools":"Role charter (ROLE)","chapter":21,"search":"role"},{"id":"F2","text":"Competence is recorded without scoring individuals, and the craft is kept up through practice.","tools":"Competence record (CMP)","chapter":21,"search":"competence"},{"id":"F3","text":"Where AI directs people, a named human decides, decisions are recorded and they can be contested.","tools":"Management decision record (MDR); contestation record (CNT)","chapter":22,"search":"contest"},{"id":"F4","text":"Employee representatives are involved before monitoring or AI-directed work starts, and impact tests are run with them.","tools":"Impact test with representatives","chapter":22,"search":"disparate"},{"id":"F5","text":"Claims about threats and the future are graded on an evidence ladder and labelled observable, expected or speculation.","tools":"Evidence record; threat register","chapter":23,"search":"threat register"}]},{"id":"G","title":"Toolbox management","chapters":"Chapters 1 to 3 and 24","practices":[{"id":"G1","text":"There is an inventory of the tools in use, with kind, owner and the book and chapter that defines each.","tools":"Tool inventory","chapter":3,"search":"inventory"},{"id":"G2","text":"Identifier families are used consistently, and one identifier means one thing.","tools":"Identifier families","chapter":2,"search":"identifier"},{"id":"G3","text":"Adoption passes a gate with evidence: a pilot, an owner and a review date.","tools":"Adoption gate","chapter":3,"search":"adoption"},{"id":"G4","text":"Tools are reviewed on a cadence, and tools that are unused or cannot be verified are retired.","tools":"Retirement criteria; toolbox review","chapter":3,"search":"retire"},{"id":"G5","text":"The toolbox is measured (evidence age, adoption, retirements) and reported to a named forum.","tools":"Toolbox measures","chapter":24,"search":"review"}]}],"chaptermap":{"1":["01_introduction","Introduction: A Toolbox for the Series","The Toolbox"],"2":["02_the_anatomy_of_a_tool","The Anatomy of a Tool: Kinds, Identifiers and Lifecycle","The Toolbox"],"3":["03_choosing_adopting_and_retiring_tools","Choosing, Adopting and Retiring Tools","The Toolbox"],"4":["04_architecture_and_decision_tools","Architecture and Decision Tools","Architecture and Delivery"],"5":["05_policy_pipeline_and_evidence_tools","Policy, Pipeline and Evidence Tools","Architecture and Delivery"],"6":["06_platform_and_infrastructure_tools","Platform and Infrastructure Tools","Architecture and Delivery"],"7":["07_inventory_classification_and_risk_tools","Inventory, Classification and Risk Tools","Governing AI"],"8":["08_obligation_assurance_and_maturity_tools","Obligation, Assurance and Maturity Tools","Governing AI"],"9":["09_supplier_contract_and_sovereignty_tools","Supplier, Contract and Sovereignty Tools","Governing AI"],"10":["10_threat_modelling_and_adversary_frameworks","Threat Modelling and Adversary Frameworks","Securing and Testing"],"11":["11_ai_red_teaming_and_emulation_tools","AI Red Teaming and Emulation Tools","Securing and Testing"],"12":["12_detection_investigation_and_intelligence_tools","Detection, Investigation and Intelligence Tools","Securing and Testing"],"13":["13_response_recovery_and_resilience_tools","Response, Recovery and Resilience Tools","Securing and Testing"],"14":["14_the_verification_loop","The Verification Loop: Purple, Chaos and Coverage Tools","Securing and Testing"],"15":["15_observability_and_service_objective_tools","Observability and Service Objective Tools","Operating"],"16":["16_change_restore_and_runbook_tools","Change, Restore and Runbook Tools","Operating"],"17":["17_capacity_cost_and_energy_tools","Capacity, Cost and Energy Tools","Operating"],"18":["18_mandate_delegation_and_autonomy_tools","Mandate, Delegation and Autonomy Tools","Agents and Autonomy"],"19":["19_observer_stop_path_and_evaluation_tools","Observer, Stop Path and Evaluation Tools","Agents and Autonomy"],"20":["20_defending_the_agent","Defending the Agent: Identity, Credential and Protocol Tools","Agents and Autonomy"],"21":["21_role_competence_and_career_tools","Role, Competence and Career Tools","People, Organisation and Society"],"22":["22_tools_for_managing_people_with_ai","Tools for Managing People with AI: Decision, Contestation and Care","People, Organisation and Society"],"23":["23_tools_for_weighing_evidence","Tools for Weighing Evidence: Threat Registers, Debate and Foresight","People, Organisation and Society"],"24":["24_building_a_toolbox","Building a Toolbox: Nordvik Region and Hansa Mutual","Experience and Conclusions"],"25":["25_conclusion","Conclusions: One Toolbox for the Series","Experience and Conclusions"]}}