generated from coulomb/repo-seed
Initial seeding of models, standards
This commit is contained in:
156
spec/agent_global-agent-brief.md
Normal file
156
spec/agent_global-agent-brief.md
Normal file
@@ -0,0 +1,156 @@
|
||||
# Global Agent Brief: InfoTechCanon
|
||||
|
||||
## Purpose
|
||||
|
||||
InfoTechCanon is a markdown-first canon for building interoperable, adaptable, and extensible information-processing systems.
|
||||
|
||||
Use it as a semantic reference layer when creating or modifying repositories, standards, schemas, profiles, mappings, integration artifacts, and agent workflows.
|
||||
|
||||
---
|
||||
|
||||
## Core Rule
|
||||
|
||||
```text
|
||||
Do not invent local semantics when a canonical concept exists.
|
||||
Import, map, profile, or extend explicitly.
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Repository Classification
|
||||
|
||||
```text
|
||||
Kernel
|
||||
Defines how the canon works.
|
||||
|
||||
Models
|
||||
Define broad domain structures.
|
||||
|
||||
Standards
|
||||
Define cross-cutting conventions or named analytical/design frameworks.
|
||||
|
||||
Profiles
|
||||
Constrain standards and models for concrete use cases.
|
||||
|
||||
Patterns
|
||||
Explain recurring practical solutions.
|
||||
|
||||
Mappings
|
||||
Relate InfoTechCanon concepts to external standards, products, or frameworks.
|
||||
|
||||
Assimilation
|
||||
Analyzes external knowledge and produces mappings, gaps, conflicts, and proposed changes.
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Current Kernel
|
||||
|
||||
```text
|
||||
InfoTechCanonCore
|
||||
InfoTechCanonKernelMap
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Current Models
|
||||
|
||||
```text
|
||||
Information Space
|
||||
Landscape
|
||||
Organization
|
||||
Governance
|
||||
Task
|
||||
Access Control
|
||||
Security
|
||||
Data
|
||||
DevSecOps
|
||||
Network
|
||||
Observability
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Current Standards
|
||||
|
||||
```text
|
||||
Tagging
|
||||
CARING Access Governance
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Do
|
||||
|
||||
- Use canonical names where available.
|
||||
- Preserve concept ownership.
|
||||
- Add mappings instead of forcing external terminology.
|
||||
- Use profiles instead of local forks.
|
||||
- Keep domain boundaries explicit.
|
||||
- Record open questions.
|
||||
- Add provenance and rationale for important changes.
|
||||
- Prefer structured front matter in Markdown artifacts.
|
||||
- Keep agent briefs aligned with full standards.
|
||||
- Treat CARING as a specialized standard, not a simple profile.
|
||||
|
||||
---
|
||||
|
||||
## Do Not
|
||||
|
||||
- Do not redefine concepts owned by another standard.
|
||||
- Do not collapse organization roles, access roles, and CARING canonical roles.
|
||||
- Do not treat tags as a substitute for fields, relationships, policies, or evidence.
|
||||
- Do not treat external standards as internal authorities.
|
||||
- Do not treat generated views as canonical source.
|
||||
- Do not hide mappings in prose.
|
||||
- Do not accept duplicated concept definitions without a boundary review.
|
||||
|
||||
---
|
||||
|
||||
## Common Distinctions
|
||||
|
||||
```text
|
||||
Actor != Subject != Principal
|
||||
|
||||
Organization Role != AccessRole != CARING Canonical Role
|
||||
|
||||
Policy != Control != Evidence
|
||||
|
||||
Option != Task != Action
|
||||
|
||||
Dataset != DataStore
|
||||
|
||||
Artifact != Deployment != Release
|
||||
|
||||
Network Intent != Network Policy != Network Configuration != Observed State
|
||||
|
||||
Alert != Incident
|
||||
|
||||
Tag != Field != Relationship
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## CARING Reminder
|
||||
|
||||
CARING analyzes access through orthogonal dimensions:
|
||||
|
||||
```text
|
||||
Subject
|
||||
Organization Relation
|
||||
Canonical Role
|
||||
Scope
|
||||
Plane
|
||||
Capability
|
||||
Exposure Mode
|
||||
Condition
|
||||
Lifecycle State
|
||||
Restriction
|
||||
Exposure Event
|
||||
Declared Access
|
||||
Effective Access
|
||||
Derived Capability
|
||||
Induced Access
|
||||
```
|
||||
|
||||
Use CARING especially for access-governance analysis, role decomposition, tenant-boundary review, agent access, service-account access, and induced capability analysis.
|
||||
Reference in New Issue
Block a user