zone-engine/SCOPE.md

102 lines
4.5 KiB
Markdown
Raw Normal View History

# SCOPE
> What this repository is about, when it is relevant, and when it is not.
> Direction lives in `GOAL.md`.
---
## One-liner
Authority for security zones — the bands of enforcement rigidity that scope
*where* a control is enforced, and the lifecycle of time-boxed exceptions that
relax them during deep refactors without leaving a permanent hole.
---
## Why this exists
Enforcement controls in this estate have been repo-wide booleans. ops-warden's
flex-auth pre-sign gate (`policy.enabled` with `fail_closed: true`) was the
first to become flippable, and flipping it would have made flex-auth a hard
dependency of every `warden sign` — including the SSH certificates the
ops-bridge tunnels depend on, one of which carries the policy call itself.
Across an estate under continuous deep refactor, uniform enforcement hardens
exactly the access needed to perform the refactor. That is not security; it is
an outage with good intentions. The flip was deferred under ops-warden's
`ADR-0006`*enforcement is zone-scoped, never a global flag* — and this repo
is what that ADR defers to.
---
## In scope
- **Zone identity and membership.** What zones exist, and what puts a lane,
actor, or workload in one. Derived from declared posture where possible
rather than a fourth hand-maintained list.
- **Enforcement stance per control per zone** — enforced / advisory / exempt,
and the failure mode (fail-closed or fail-open) that goes with it. The failure
mode is the part with teeth.
- **Time-boxed exception lifecycle** — a relaxation with an expiry that is
enforced by something rather than intended, plus the record of who widened
what, when, and until when.
- **The declaration format** repos use to state their zones, following
`tenancy.yaml` and its rule: *accuracy, not altitude*.
- **Drafting the canon standard** for the model, offered to `net-kingdom` for
publication the way `tenant-engine` and `user-engine` have boundary contracts
in canon.
## Out of scope
- **Authorization decisions.** `flex-auth` is the PDP and stays the only one.
A second decision point is the split-brain this repo must not create.
- **Network segmentation.** "Zone" is overloaded; this is about enforcement
rigidity, not routing or NetworkPolicy. The two must not be conflated.
- **Substrate placement.** Reefs (`reef-railiance`, `reef-storage`) answer
*where does it run*; a zone answers *how rigidly is it governed*. Correlated,
not the same axis — and per repo-manager's rule, topology is not readiness.
- **Identity, secrets, tenancy.** `key-cape`, OpenBao / `secrets-engine`,
`tenant-engine`.
- **Implementing any specific control.** Owners keep their own controls; this
repo says where they bite.
- **Publishing canon.** `net-kingdom` owns canon; `policy-nexus` publishes.
---
## Relationship to what already exists
| Mechanism | Owner | Relationship |
| --- | --- | --- |
| `tenancy-posture_v0.1` (six axes, per-repo `tenancy.yaml`) | net-kingdom canon | **Structural model to follow** — graduated levels, declared conformance, review dates, "accuracy not altitude". Its §14 stance is *structure, not tooling*, which is why this repo starts as a model and not an API |
| Environment posture + workload maturity `M0``M3` | ops-warden (WP-0015) | Inputs a zone may derive membership from |
| `organization_posture: build` | ops-warden (WP-0029) | Candidate to fold in rather than run alongside |
| Reefs / `bound_reefs` | repo-manager | Correlated substrate axis; must be reconciled, not duplicated |
| Compiled registry snapshot | flex-auth | **How zone membership reaches the PDP.** flex-auth is latency-critical; compilation, not runtime lookup |
| `warden plan` verdicts + `reasons` | ops-warden (WP-0029) | Existing verdict machinery to extend, not parallel |
---
## Current state (2026-08-19)
Seeded. No model, no schema, no runtime. `ZONE-WP-0001` derives the model from
the real estate and drafts the canon standard; a runtime is earned only if the
exception lifecycle proves to need one.
Deliberately empty of schema — an empty engine that ships a schema first ends up
defining the model by accident.
---
## Relevant when
- A control is about to be enabled and the question "enforced *where*?" has no
answer
- A deep refactor needs relaxed rigidity in a band of the estate, with an expiry
- A repo needs to declare which zone its lanes sit in
## Not relevant when
- Asking whether a specific request is allowed (→ `flex-auth`)
- Placing a workload on a substrate (→ `repo-manager` reefs)
- Anything about network reachability