Implement canon conformance and maintenance optimizations
All checks were successful
CI Smoke / host-smoke (push) Successful in 0s
CI Smoke / container-smoke (push) Successful in 3s

Assistant: codex
Assistant-Model: gpt-6-astra
Assistant-Session: 01a06e82-3e08-7042-a79d-438ac6eed8db
This commit is contained in:
tegwick 2026-09-05 00:50:09 +02:00
parent a2e7f22d8d
commit b081d39da1
64 changed files with 4491 additions and 373 deletions

308
SCOPE.md
View file

@ -1,241 +1,115 @@
# SCOPE
## Current Repository Scope
Reviewed: 2026-09-05 (Europe/Berlin), including the uncommitted working tree.
This repository currently contains the seed structure for **InfoTechCanon**: an evolving, markdown-first canon for building interoperable, adaptable, and extensible information-processing systems.
## Current repository scope
At this stage, the repository is not yet a fully validated standards system. It is a **seed corpus** of kernel documents, domain models, specialized standards, and integration guidance that together define the first-generation InfoTechCanon kernel.
InfoTechCanon maintains one concrete, Markdown-first, machine-readable canon
under `infospace/`, with a Python service for inspection, validation, retrieval,
and generation. It has progressed beyond its seed corpus to a **service baseline
with executable proofs and mixed artifact maturity**.
---
The working-tree canon version is `0.7.0`; the Python package version is
`0.1.0`. These are separate version surfaces. Many models remain `RC1-seed`;
ITC-CAP is a frozen draft at `0.4.0`, and Emission Cadence is a draft document
at `0.1.0` with wire schema version `0.1`. This is not yet a stable,
ecosystem-wide conformance release.
## Current Purpose
## Ownership and boundaries
The repository currently serves to:
Following [INTENT.md](INTENT.md), this repository occupies the Taxonomy layer:
it owns shared terminology, generic semantic contracts, compatibility rules,
and reference artifacts for independently owned systems.
- collect the first generation of InfoTechCanon standards and models,
- define the semantic boundaries between domains,
- preserve orthogonality between concepts,
- provide seed structures for mappings, profiles, patterns, assimilation, validation, and agent use,
- support future machine-readable extraction,
- and act as the first major use case for markdown-based infospaces and related tooling.
The CLI, library, and local read-only HTTP API serve and check those artifacts.
They do not establish ownership of consumer runtime state, enforcement,
observation, or orchestration. NetKingdom owns its security architecture and
importing security profiles. Sources own emission-cadence instances; observers
own evaluation and findings; importing profiles own additional policy requirements.
---
## Implemented canon content
## Current Kernel
| Surface | Current responsibility |
| --- | --- |
| `infospace/kernel/` | Core mechanisms for artifacts, ownership, relationships, mappings, assimilation, profiles, patterns, conformance, provenance, and interfaces; kernel map. |
| `infospace/models/` | Information Space, Landscape, Organization, Governance, Task, Access Control, Security, Data, DevSecOps, Network, Observability, and Capability models. |
| Extensions and catalogs | PURPOSES/Purpose and Demand governance extension; EvidenceBasis; AttributeValueType; ITC-CAP capabilities, requirements, provisions, maturity, economics, and evidence relationships. |
| `infospace/standards/` | Tagging, CARING access governance, Repository Layout, and draft Emission Cadence. |
| `infospace/profiles/small-saas/` | Executable example linking service, system, tenants, user, team, data, deployment, task, policy, control, evidence, incident, and capability requirements/provisions. |
| `infospace/patterns/` | IntentScopePurposes, PracticePattern scheme, InterfaceDeprecationStrangler, and AgenticDrivesFunctional. |
| Mappings and assimilation | Mapping artifacts and source-preserving workspaces for IT Capability Canon and Emission Cadence, including comparisons, dispositions, gaps, and open questions. |
| `infospace/evaluations/` | Evaluation/comparison packs for user-engine, railiance-fabric, repo-scoping, and repository-layout. These prepare adoption and review; they do not certify deployed consumers. |
| CARING Kubernetes RBAC benchmark | Native-concept mappings, structured access descriptors, findings, conflicts, and proposed improvements. |
| `infospace/agent/` | Generated retrieval indexes and artifact briefs, consumer briefs, interface-card template, alignment review kit, and consumer workplan template. |
The current kernel consists of:
The root is the service and governance shell. `seeds/` preserves provenance.
`incoming/`, `demand/`, and `feedback/` distinguish source material, consumer
needs, and utility evidence. Workplans and intake records track work;
`history/` holds dated assessments. Generated navigation lives under
`infospace/indexes/`, `infospace/views/`, and `infospace/agent/`.
```text
kernel/
InfoTechCanonCore.md
InfoTechCanonKernelMap.md
```
## Implemented service and checks
### InfoTechCanonCore
`src/info_tech_canon/` provides an importable Python service, JSON-first CLI,
and read-only HTTP API, defaulting to `127.0.0.1:8765`. Operations include
artifact/model/standard inspection, graph export (JSON or Mermaid), canon and
profile validation, capability-record review, generated-view retrieval, and
alignment-review materials. CLI maintenance commands generate indexes, trees,
and agent briefs. Usage and endpoints are in [README.md](README.md).
Defines shared canon mechanisms:
Validation checks registered paths, artifact IDs and relationship targets,
collection metrics, required schema presence, and selected content contracts.
Specialized checks cover capability records/catalogs, practice patterns,
consumer packs, and the built-in Emission Cadence example. JSON Schemas exist
for multiple artifact types, but general schema-to-instance validation and
semantic conformance are not complete across the corpus.
- canon artifacts,
- concepts,
- concept ownership,
- relationships,
- mappings,
- assimilation,
- profiles,
- patterns,
- validation rules,
- conformance levels,
- versioning,
- provenance,
- agent briefs,
- and Canon Interface Cards.
Dependencies are Python 3.12+, PyYAML, jsonschema, and `infospace-bench`.
The loader discovers an installed bench distribution. An explicit `--root` or
`INFO_TECH_CANON_ROOT` selects a separately supplied corpus. Narrow wheel
installation is supported; upstream bench's full transitive dependency metadata
still contains workstation-specific paths. See
[consumption instructions](docs/canon-consumption.md).
### InfoTechCanonKernelMap
## Verified baseline and limits
Shows how the first-generation standards and models fit together, including the position of CARING as a specialized access-governance standard.
After INFO-WP-0019 implementation, `make check` passed **36 tests** and
verified **77 generated assets**. Read-only CLI validation
reported **76 registered artifacts, 315 relationships, zero errors and zero
warnings**. Small-SaaS validation passed with **15 artifacts/payloads**.
---
The ownership calculation reports 129 entries and no cross-owner conflicts,
but derives entries from artifact titles and explicit `owned_concepts`
frontmatter. It does not establish ownership of every concept in prose.
Collection coverage metrics measure the registered corpus, not fulfillment
of every intent commitment.
## Current Models
Current limitations include:
The repository currently includes seed models for:
- incomplete concept extraction and enforcement of concept-level imports;
- uneven schema depth and validation coverage, including no formal
CARINGAccessDescriptor schema or complete effective-access calculus;
- limited external mapping and independent consumer interoperability evidence;
- interface templates and evaluation packs without a complete set of
source-owned, version-pinned consumer declarations;
- draft promotion and compatibility questions, including the requirement for
two independent Emission Cadence source implementations;
- upstream full-application packaging limitations and consumer-owned adoption evidence.
```text
models/
information-space/
landscape/
organization/
governance/
task/
access-control/
security/
data/
devsecops/
network/
observability/
```
INFO-WP-0019 adds reusable arbitrary Emission Cadence declaration validation,
mapping schema checks, explicit ownership-conflict enforcement, a validation
coverage report, content-addressed contract export, generated-asset freshness
checks, scope inventory, and validation source/time evidence. These mechanisms
do not close the semantic or independent-adoption gaps listed above.
These models define broad domain structures.
## Exclusions and evolution
| Model | Current Responsibility |
|---|---|
| Information Space | Markdown-first knowledge packaging, retrieval, links, chunks, indexes, agent briefs |
| Landscape | IT landscapes, services, systems, runtime resources, infrastructure context |
| Organization | Actors, roles, teams, membership, responsibility, authority, accountability |
| Governance | Policies, rules, decisions, controls, risk, evidence, exceptions, assurance |
| Task | Work items, options, tasks, actions, blockers, dependencies, commitment, outcomes |
| Access Control | Subjects, principals, permissions, grants, authorization decisions, enforcement |
| Security | Threats, weaknesses, vulnerabilities, findings, exposure, attack paths, incidents |
| Data | Datasets, schemas, data products, classification, lineage, quality, contracts |
| DevSecOps | Source-to-artifact-to-release-to-deployment delivery flow and evidence |
| Network | Addressing, topology, routing, policy, reachability, exposure, network state |
| Observability | Telemetry, logs, metrics, traces, alerts, SLOs, health, operational evidence |
This repository does not replace external standards, mandate one universal
enterprise model, operate consumer systems, enforce NetKingdom policy, or
provide production observability or authorization services. A complete
ontology and exhaustive external-standard mappings are not current deliverables.
---
## Current Specialized Standards
The repository currently includes specialized standards for:
```text
standards/
tagging/
caring/
```
| Standard | Current Responsibility |
|---|---|
| Tagging | Tag identity, schemes, namespaces, assignments, mappings, validation |
| CARING | Orthogonal access-governance analysis using canonical roles, planes, exposure modes, derived and induced capabilities |
---
## Current Structural Assumptions
The current repository structure assumes the distinction:
```text
Kernel
Defines how the canon works.
Models
Define broad domain structures.
Standards
Define cross-cutting conventions, mechanisms, or named analytical/design frameworks.
Profiles
Constrain models and standards for concrete implementation contexts.
Patterns
Describe recurring practical solutions.
Mappings
Relate InfoTechCanon concepts to external bodies of knowledge.
Assimilation
Analyzes external knowledge bodies and turns them into mappings, gaps, conflicts, and proposed changes.
Views
Provide generated or curated navigation over canon artifacts.
Agent
Provides compact, retrieval-optimized guidance for AI agents and tools.
```
---
## Current Non-Goals
The current repository does not yet provide:
- a complete ontology,
- complete YAML/JSON schemas for all artifact types,
- automated validation tooling,
- fully extracted concept pages,
- finalized mappings to external standards,
- complete application profiles,
- complete agent retrieval indexes,
- mature conformance tests,
- or a stable release of the InfoTechCanon.
The current scope is seed-level structure and consolidation.
---
## Current Known Duplication
The current seed standards still repeat some generic sections that should eventually be centralized in `InfoTechCanonCore`, including:
- mapping model,
- assimilation model,
- profile format,
- conformance levels,
- validation rule format,
- repository placement,
- agent brief requirements,
- Canon Interface Card usage,
- lifecycle statuses.
This duplication is accepted for the seed phase but should be reduced during refactoring.
---
## Current Refactoring Priorities
Near-term work should focus on:
1. moving repeated generic mechanisms into Core,
2. creating machine-readable schemas,
3. extracting concept pages from seed standards,
4. creating global indexes,
5. creating agent briefs,
6. creating first application profiles,
7. creating formal assimilation workspaces,
8. running CARING as the first full benchmark assimilation,
9. validating concept ownership across the kernel,
10. and creating Canon Interface Cards for initial tools and repositories.
---
## First Validation Targets
The first validation targets should be:
```text
1. Concept ownership table
2. Cross-standard import matrix
3. Kernel dependency graph
4. CARING access descriptor schema
5. Mapping schema
6. Profile schema
7. Agent brief schema
8. Repository layout validation
9. Internal link validation
10. Duplicate concept detection
```
---
## Current Maturity
The repository is currently at:
```text
Maturity: Seed Kernel
Status: Pre-release consolidation
Primary use: Human/agent reference and refactoring foundation
```
The next maturity target is:
```text
Maturity: Kernel RC1
Condition:
Core mechanisms centralized,
standards indexed,
concept ownership explicit,
first profiles and mappings created,
CARING assimilation completed,
and basic validation scripts available.
```
Further semantic coverage, stronger conformance, and consumer adoption are
consistent with intent but require follow-up work. Evidence, gaps, and
optimization options are recorded in the
[2026-09-05 scope assessment](history/2026-09-05_002437+0200-scope-against-intent.md).