Files
tom_analysis_agent/eval/results/20261021T203623Z-handoff/operating_model/scope.yaml
Jonathan Boniface 628b6ee170 docs: publish the runs of 2026-10-21
The evaluation (both cases three times: report, results, spread, each
run's Deliverables and lesson candidates) and the hand-off through the
Orchestration engine (report, the document of links, both agents' kept
files and records). Benchmark cases only, all fictional; no machine paths,
and the links carry no signatures. Other runs stay ignored.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-22 09:25:03 +01:00

467 lines
14 KiB
YAML

schema_version: 1
metadata:
customer_case: case-08-landing-zone
client_name: ask
author: ask
revision: ask
source: architecture:6c316f05-9643-492c-bab1-eed56e108754
status: UNDER_REVIEW
engagement:
kind: platform_foundation
intent:
statement: We are a UK-regulated payment platform expanding into Google Cloud. We
need an enterprise-grade landing zone ready in under 10 weeks to support our core
payments engine. All payment processing must comply with DORA and PCI-DSS v4.0.
We have existing on-premises systems that require private, low-latency connectivity
over a 10Gbps dedicated interconnect. We use Entra ID for corporate single sign-on
and GitHub Actions for our delivery pipelines. Security requires that all egress
traffic flows through centralised inspection appliances before touching the public
internet, and customer data must be encrypted with keys we control.
requirements: ask
business_outcomes:
- outcome: Continuous and resilient payment processing with high availability and
low-latency hybrid connectivity
id: BO-1
- outcome: Sustained regulatory compliance with DORA and PCI-DSS v4.0 security and
resilience standards
id: BO-2
- id: BO-3
outcome: Rapid and secure platform change releases through automated CI/CD delivery
pipelines
exclusions:
- Application container deployment and runtime microservices managed by application
squads
- Active-active secondary region in phase 1 (warm standby adopted)
solution:
cloud_provider: gcp
items:
- id: S-1
name: Multi-Tier Environment Hierarchy & Isolation Boundary
kind: governance
product: Resource Manager
source_ref: C-1
elements:
- E-1
- E-2
- E-3
- id: S-2
name: Dedicated Hybrid Network Interconnect Gateway
kind: network
product: Cloud Interconnect
source_ref: C-2
elements:
- E-9
- E-10
- E-11
- E-12
- E-35
- E-36
- id: S-3
name: Centralised Egress Traffic Inspection Gateway
kind: compute
product: Compute Engine
source_ref: C-3
elements:
- E-13
- E-14
- E-15
- E-16
- E-17
- id: S-4
name: Transit Network Fabric & Routing Controller
kind: service
product: Virtual Private Cloud
source_ref: C-4
elements:
- E-4
- E-5
- E-40
- E-41
- E-42
- E-6
- E-7
- E-8
- E-43
- E-44
- E-45
- E-37
- E-47
- E-51
- E-52
- E-53
- id: S-5
name: Corporate Identity Federation Gateway
kind: identity
product: Cloud Identity
source_ref: C-5
elements:
- E-18
- id: S-6
name: Automated Delivery Pipeline Identity Broker
kind: identity
product: Workload Identity Federation
source_ref: C-6
elements:
- E-19
- E-20
- E-54
- E-55
- id: S-7
name: Cryptographic Key Management & Rotation Service
kind: keys
product: Cloud KMS
source_ref: C-7
elements:
- E-21
- E-22
- E-23
- E-24
- id: S-8
name: Centralised Immutable Audit Log Repository
kind: storage
product: Cloud Storage
source_ref: C-8
elements:
- E-25
- E-26
- id: S-9
name: Platform Compliance & Governance Guardrail Controller
kind: governance
product: Organization Policy Service
source_ref: C-9
elements:
- E-27
- E-28
- E-29
- E-30
- E-38
- E-46
- E-56
- E-57
- id: S-10
name: Declarative Infrastructure Provisioning & State Engine
kind: storage
product: Cloud Storage
source_ref: C-10
elements:
- E-31
- id: S-11
name: Centralised Operational Telemetry & Alerting Engine
kind: observability
product: Cloud Monitoring
source_ref: C-11
elements:
- E-32
- E-33
- E-58
conditions:
- id: K-1
'on':
- S-1
- S-7
- S-9
condition: Customer data and payment processing remain within United Kingdom and
European Union legal jurisdictions.
source_ref: NFR-1
- id: K-2
'on':
- S-1
- S-7
- S-9
condition: Warm standby environment operates with reduced capacity until activated
during a disaster recovery failover.
source_ref: NFR-1
- id: K-3
'on':
- S-1
- S-9
condition: PCI-regulated production environment is isolated from non-PCI scopes
according to PCI-DSS v4.0 network segmentation requirements.
source_ref: NFR-2
- id: K-4
'on':
- S-3
- S-4
condition: Workload APIs and payment services are accessible exclusively through
private corporate network paths.
source_ref: NFR-3
- id: K-5
'on':
- S-2
condition: Hybrid connection maintains low-latency, SLA-backed private data transmission
between on-premises and Google Cloud.
source_ref: NFR-4
- id: K-6
'on':
- S-3
condition: Centralised inspection appliances perform next-generation stateful
packet inspection, URL filtering, and threat detection on outbound internet
sessions.
source_ref: NFR-5
- id: K-7
'on':
- S-2
- S-4
condition: Platform foundation failover to europe-west1 can be completed within
4 hours during a primary region outage.
source_ref: NFR-6
- id: K-8
'on':
- S-2
- S-4
condition: Configuration drift and backup state achieve a maximum data loss window
(RPO) of 1 hour.
source_ref: NFR-6
- id: K-9
'on':
- S-5
- S-6
condition: Corporate user lifecycle management and MFA enforcement are governed
centrally by Entra ID conditional access policies.
source_ref: NFR-7
- id: K-10
'on':
- S-1
- S-9
condition: Platform governance and resilience architecture comply with DORA regulatory
testing and third-party risk management frameworks.
source_ref: NFR-8
- id: K-11
'on':
- S-1
- S-9
condition: Production cardholder data environment satisfies PCI-DSS v4.0 isolation
and security control requirements.
source_ref: NFR-8
- id: K-12
'on':
- S-7
condition: Confidential payment data is secured against unauthorized platform-level
access or data exposure.
source_ref: NFR-9
- id: K-13
'on':
- S-8
condition: Audit logs cannot be modified, deleted, or shortened by any identity
or administrator prior to retention expiration.
source_ref: NFR-10
- id: K-14
'on':
- S-2
- S-4
- S-11
condition: Monthly availability of the core platform landing zone and network
routing meets or exceeds 99.9%.
source_ref: NFR-11
- id: K-15
'on':
- S-9
condition: Primary and secondary regions leverage Google Cloud data centres with
published carbon-free energy (CFE) performance metrics.
source_ref: NFR-12
- id: K-16
'on':
- S-6
- S-10
condition: Platform foundation delivery and workload ingestion readiness are achieved
within 10 weeks of project commencement.
source_ref: NFR-13
- id: K-17
'on':
- S-8
- S-10
- S-11
condition: Central platform team operates 24/7 support for infrastructure incidents
and availability alerts.
source_ref: NFR-14
- id: K-18
'on':
- S-8
- S-10
- S-11
condition: Pull request workflows require automated CI/CD linting, validation,
and peer approval prior to merge and deployment.
source_ref: NFR-14
- id: K-19
'on':
- S-3
condition: NVA instances use machine families that support high bandwidth networking.
source_ref: DR-3
- id: K-20
'on':
- S-3
condition: The warm standby environment operates at a minimal pilot light footprint
during normal operations.
source_ref: DR-11
- id: K-21
'on':
- S-3
condition: Automated disaster recovery scripts scale the standby environment to
full capacity only during a failover event.
source_ref: DR-11
build:
duration_days: 50
customer_owned:
- Order and provision the dedicated carrier circuit
inherited_risks: ask
operations:
support_hours: 24x7
availability_hours: 24x7
change_process: Infrastructure changes managed via pull requests with required approvals
and automated GitHub Actions CI/CD pipelines
monitoring: ask
sourcing:
model: provider_operated
delivery_countries:
- GB
- PL
- IN
organisation:
existing_teams: []
headcount_limit: ask
service_management:
tooling: Google Cloud Operations Suite (Cloud Monitoring and Cloud Logging)
service_levels:
priority_1_response_minutes: 30
recovery_time_hours: 4
compliance:
frameworks:
- DORA
- ISO27001
- PCI_DSS_v4
commercial:
currency: GBP
monthly_run_budget: ask
transition:
go_live_date: ask
run:
pre_approved: true
overlay_skills: ask
inferred_assumptions:
- id: ASSUMP-01
field: intent.business_outcomes
statement: 'Assumed intent.business_outcomes is [{''outcome'': ''Continuous and
resilient payment processing with high availability and low-latency hybrid connectivity'',
''id'': ''BO-1''}, {''outcome'': ''Sustained regulatory compliance with DORA and
PCI-DSS v4.0 security and resilience standards'', ''id'': ''BO-2''}, {''id'':
''BO-3'', ''outcome'': ''Rapid and secure platform change releases through automated
CI/CD delivery pipelines''}]. Derived from customer request for a UK-regulated
payment platform requiring high resilience, continuous compliance, and automated
delivery.'
impact_if_invalid: Changes what is measured, and what each measure is for.
- id: ASSUMP-02
field: intent.exclusions
statement: Assumed intent.exclusions is ['Application container deployment and runtime
microservices managed by application squads', 'Active-active secondary region
in phase 1 (warm standby adopted)']. Derived from workload scope boundaries where
application workloads are managed by application squads, and secondary region
is warm standby rather than active-active.
impact_if_invalid: Changes which Capabilities are designed.
- id: ASSUMP-03
field: solution.build.duration_days
statement: Assumed solution.build.duration_days is 50. Derived from 10-week delivery
window at 5 working days per week.
impact_if_invalid: Changes when Capabilities must be ready.
- id: ASSUMP-04
field: operations.change_process
statement: Assumed operations.change_process is Infrastructure changes managed via
pull requests with required approvals and automated GitHub Actions CI/CD pipelines.
Derived from request and scope specification indicating pull-request driven changes
with peer review and GitHub Actions CI/CD automation.
impact_if_invalid: Changes change enablement and release.
- id: ASSUMP-05
field: organisation.existing_teams
statement: 'Assumed organisation.existing_teams is []. Catalogue default: no existing
operational teams assumed; roles to be provisioned under provider-operated model.'
impact_if_invalid: Changes which Roles are new and which are filled.
- id: ASSUMP-06
field: service_management.tooling
statement: 'Assumed service_management.tooling is Google Cloud Operations Suite
(Cloud Monitoring and Cloud Logging). Catalogue default: Google Cloud Operations
Suite (Cloud Monitoring and Cloud Logging) with no separate third-party ticketing
system mandated.'
impact_if_invalid: Changes tooling and the processes written for it.
- id: ASSUMP-07
field: service_levels.priority_1_response_minutes
statement: Assumed service_levels.priority_1_response_minutes is 30. Catalogue default
for Priority 1 incident response acknowledgement.
impact_if_invalid: Changes on-call arrangements and staffing.
- id: ASSUMP-08
field: service_levels.recovery_time_hours
statement: Assumed service_levels.recovery_time_hours is 4. Catalogue default and
aligned with target 4-hour RTO for warm standby disaster recovery in secondary
region.
impact_if_invalid: Changes continuity arrangements.
discovery_gaps:
- field: metadata.client_name
prompt: What is metadata.client_name?
severity: LOW
- field: metadata.author
prompt: What is metadata.author?
severity: LOW
- field: metadata.revision
prompt: What is metadata.revision?
severity: LOW
- field: intent.requirements
prompt: Which requirements for operations has the customer stated outright?
severity: LOW
- field: solution.inherited_risks
prompt: What is already known to be unverified or at risk in the solution?
severity: LOW
- field: operations.monitoring
prompt: Which monitoring and alerting is already in place or required?
severity: LOW
- field: organisation.headcount_limit
prompt: Is there a limit on the number of people operating this?
severity: LOW
- field: commercial.monthly_run_budget
prompt: Is there a monthly budget for operating this, in that currency?
severity: LOW
- field: transition.go_live_date
prompt: When must the solution be in operation?
severity: LOW
- field: run.overlay_skills
prompt: Which overlay Skills should this Execution load?
severity: LOW
provenance:
commercial.currency: digested
commercial.monthly_run_budget: unknown
compliance.frameworks: digested
engagement.kind: digested
intent.business_outcomes: assumed
intent.exclusions: assumed
intent.requirements: unknown
intent.statement: digested
metadata.author: unknown
metadata.client_name: unknown
metadata.customer_case: digested
metadata.revision: unknown
metadata.source: digested
operations.availability_hours: digested
operations.change_process: assumed
operations.monitoring: unknown
operations.support_hours: digested
organisation.existing_teams: assumed
organisation.headcount_limit: unknown
run.overlay_skills: unknown
run.pre_approved: supplied
service_levels.priority_1_response_minutes: assumed
service_levels.recovery_time_hours: assumed
service_management.tooling: assumed
solution.build.customer_owned: digested
solution.build.duration_days: assumed
solution.cloud_provider: digested
solution.conditions: digested
solution.inherited_risks: unknown
solution.items: digested
sourcing.delivery_countries: supplied
sourcing.model: supplied
transition.go_live_date: unknown
readiness:
supplied: 3
answered: 0
assumed: 8
unknown: 10
not_applicable: 0
digested: 12