rfc-app/docs/design/2026-06-06-configurable-collection-metadata.md

Solution Design: Configurable Collection Metadata (clean-doc tagging)

Author(s)	Ben Stull
Reviewers / approvers	Ben Stull
Status	draft
Version	v0.1.2
Source artifacts	Reference modeled: retired BDD Release Planner (wiggleverse/wiggleverse-ecomm-bdd-release-planner-app, RETIRED 2026-06-04) · Related spec: docs/design/2026-06-05-three-tier-projects-collections.md (§22) · Corpus: ecomm Shopify-modeled BDD (wiggleverse-ecomm-meta/research/shopify, ~1,238 scenarios) · Supersedes: —

Change log

Date	Version	Change	By
2026-06-06	v0.1.0	Initial draft from discovery session OHM-0079.0	Ben Stull
2026-06-06	v0.1.1	Executive Summary → value-only; add §4 Pain Points; renumber per Solution Design template revision	Ben Stull
2026-06-06	v0.1.2	Reshuffle (Business Outcomes → §5, business cluster); outcomes restated as business (adoption/diversity), not solution outputs; Business Use Cases rewritten as solution-agnostic actor-goal scenarios (no product reference)	Ben Stull

---

1. Executive Summary

A deployment's corpus is only as valuable as the ability of the people running it to prioritise it, navigate it, and act on it — and as valuable as the downstream tools that can read structured signal out of it. Today that value is stranded: operators and contributors can't rank what matters or find content by what matters, and the tools meant to plan and build from the corpus have nothing structured to consume. This design delivers, for those people, the ability to rank what matters, find content by what matters, and hand clean machine-readable signal to the collaborators and tools that depend on it — turning a flat thousand-item corpus into something a team can actually plan and a downstream planner can build on, while keeping the source documents clean and readable. The outcome for the customer: lower-friction corpus planning for operators and contributors, broader adoption by teams whose document types rfc-app couldn't previously serve, and a corpus external tooling can consume without bespoke glue. (Value summary; scope and mechanism begin at §7.)

2. Business Context

The framework hosts RFC standardization for multiple deployments. One deployment hosts the ecomm BDD corpus — ~1,238 Shopify-modeled scenarios, one markdown file per scenario, slugged by feature ID (DD-FF-NNNN-slug). A standalone BDD Release Planner previously let operators search that corpus, attach metadata (priority P0–P3, owner, status), cluster scenarios into named releases, and emit each release as a roadmap phase. §22 (three-tier projects/collections) absorbed the planner's corpus hosting into rfc-app (the corpus now runs as a bdd project on the RFC deployment) and the planner was retired — but its annotation half (priority/tags on scenarios, filtering, bulk assignment) was never rebuilt. Operators planning work, and downstream tools consuming the corpus, currently have no structured, framework-native way to set or read that metadata. More broadly, teams evaluating rfc-app for other document types often need structured attributes (a priority, a status, domain tags) the platform can't yet express — so they go elsewhere.

3. Problem Statement

rfc-app cannot express or surface structured signal about its content. Tags are free-form strings with no filtering; there is no notion of priority or any other collection-defined attribute; the catalog is a flat list; and what little metadata exists is mixed into the top of every document. As a result, a corpus cannot be prioritised, navigated by attribute, planned in bulk, or cleanly consumed by downstream tools — and teams whose workflows depend on such attributes cannot adopt the platform at all.

4. Pain Points

#	Pain	Who feels it	Cost / frequency today
PP-1	Scenarios carry no priority, so triage and planning happen off-platform, in spreadsheets and memory	Release planner, contributor	Every planning cycle; signal lives off-platform and goes stale
PP-2	The catalog is a flat, unfilterable list — at ~1,200 scenarios, "show me the P0 checkout scenarios" is impractical	Reader, release planner	Every browse/triage; finding the right work is slow and error-prone
PP-3	Tags are free-form with no filtering payoff, so they're decorative and go unmaintained	Contributor	Ongoing; the one existing affordance rots
PP-4	Annotating many scenarios means opening many PRs, so bulk planning has no home in the tool	Release planner	Every batch; the core planning gesture is effectively impossible
PP-5	rfc-app metadata clutters the top of every document, hurting readability and making the corpus awkward to consume cleanly	Reader, downstream consumer	Every read; every downstream integration
PP-6	Downstream tools have no structured signal to read — the retired planner's capability left a gap	Downstream consumer	Continuous since the planner's retirement
PP-7	Teams whose document types need structured attributes can't model them, so they don't adopt rfc-app	Prospective adopter (org/team)	Every evaluation that ends in "not yet"

5. Targeted Business Outcomes

Business outcomes for rfc-app as a platform — adoption, reach, and diversity of use — not solution outputs. (Whether documents carry a priority is a solution output, tracked as a slice's Definition of Done in §13, not here.) These are the business change we expect once the pains in §4 are relieved.

Outcome	Success metric	Baseline → Target	Guardrail (must not regress)	How / when measured
Teams blocked by missing structured attributes (priority / structured tags) now adopt rfc-app	# organizations on rfc-app; # active users	internal deployments only → external orgs onboard	existing deployments don't churn	deployment registry + usage analytics; quarterly
The platform hosts a wider variety of workflows and document types	# distinct document/collection types & field schemas in use; # distinct workflow patterns	today's handful (philosophy RFCs, BDD corpus) → broader mix	existing types' experience unchanged	type/schema census across deployments; quarterly
Corpus planning happens on-platform rather than in side tools	share of prioritisation/planning done in rfc-app vs spreadsheets	largely off-platform → on-platform	—	operator interviews + usage signals; quarterly

6. Stakeholders / Personas / Actors

Actor	Type	Goal in this design
Corpus contributor	persona	Set priority and tags on a scenario when proposing/curating it
Release planner (operator)	persona	Bulk-prioritise/tag many scenarios quickly to shape a plan
Collection owner	operator	Declare the metadata fields a collection supports
Prospective adopter (org/team)	persona	Run a new document-type workflow on rfc-app
Downstream consumer	system	Read structured per-scenario metadata from the git corpus (e.g. an external release planner)
Reader	persona	Read clean scenario docs; filter the catalog by priority/tag

7. Scope

• In scope: per-collection field schema in .collection.yaml (enum, tags, text); per-entry sidecar (<slug>.meta.yaml) as source of truth with pure-prose doc bodies; schema-derived faceted left-pane filtering with counts; schema-derived metadata form (detail panel); single + bulk tag/untag with direct commit for authorized roles; dual-read compatibility + a one-shot frontmatter→sidecar migration tool + lazy migration on write.
• Out of scope: release ordering, ship status, roadmap emission, RELEASE-PLAN.md generation — these stay downstream, consuming sidecars from git. In-app management of field definitions / controlled vocabularies; corpus-wide tag rename/merge/delete. Sub-document (per-scenario-within-a-file) grain. A whole-corpus metadata export endpoint.
• Non-goals: rebuilding a bespoke "release" entity in rfc-app — releases are not modeled here at all; metadata is the only primitive.

8. Assumptions · Constraints · Dependencies

• Assumptions: git remains the content source of truth and downstream consumers can read the corpus from git; the BDD grain is one markdown file per scenario (already true for the ecomm corpus), so no sub-document parsing is needed.
• Constraints: rfc-app is a framework hosting multiple deployments — the upgrade must be mechanical and non-breaking, with §20 changelog/upgrade-steps; the hard secrets rule (§6.3) holds; metadata edits must respect scope-role authorization (§22 Part B / S3).
• Dependencies: the S3 scope-role resolver (auth.effective_scope_role) for edit authorization; the existing git write-through path used by edit-meta (§9.5); the §22 collection model (.collection.yaml, cached_rfcs keyed by (collection_id, slug)).

9. Business Use Cases

Business use cases are solution-agnostic: they state what an actor wants to accomplish, why (the value), and what success looks like — with no reference to rfc-app or any product. Each is true regardless of how it's met; in principle a person could satisfy it by hand (a concierge first pass) before any software is built, and only later is it automated. They open with "As a … I can … so that …" and state acceptance in business terms.

BUC-1 — As a release planner, I can prioritise the requirements in a body of work, so that I can decide what belongs in upcoming releases.

Scenario: BUC-1 — Prioritise to plan releases
  Given a body of requirements of varying importance
  When the planner weighs which matter most
  Then they hold a ranking of those requirements by importance
  And can decide a release's contents from it

• Acceptance: the planner can select and justify the next release's contents from the relative importance of the work.

BUC-2 — As a planner facing a large body of requirements, I can organise and triage it within a normal working session, so that planning actually gets done rather than deferred or improvised.

Scenario: BUC-2 — Triage at scale
  Given more requirements than can be weighed one at a time
  When the planner ranks and groups them in bulk
  Then the body of work reflects those decisions without per-item drudgery

• Acceptance: a planner moves from an unsorted corpus to a prioritised plan in one sitting.

BUC-3 — As a team, I want the importance and categorisation of our requirements captured durably and shareably, so that other people and tools can plan from it without re-deriving it.

Scenario: BUC-3 — Durable, shareable signal
  Given requirements that have been weighed and categorised
  When someone or something else needs to plan from them
  Then they can read what matters and why without asking the original author

• Acceptance: a second party — person or tool — can pick up the work and plan from it unaided.

BUC-4 — As a team with a specialised body of documents, I can capture the attributes that make them actionable (importance, status, category), so that I can manage that work the way my domain requires.

Scenario: BUC-4 — Manage a domain's work on its own terms
  Given documents whose usefulness depends on domain-specific attributes
  When the team records and works with those attributes
  Then they can run their workflow with the distinctions it depends on

• Acceptance: the team can capture and act on the distinctions their domain requires — success is them choosing to manage the work this way.

BUC-5 — As someone consuming a large corpus, I can find the items that matter to my current purpose, so that I act on the right things instead of wading through everything.

Scenario: BUC-5 — Find what matters
  Given a large body of items
  When the consumer looks for the important ones for their task
  Then they can locate them quickly

• Acceptance: a person narrows a large corpus to the relevant, important subset for their task.

10. Product Use Cases

Scenario: PUC-1 — Set priority/tags on a scenario (realizes BUC-1, BUC-4)
  Given I am a contributor viewing a scenario whose collection defines priority and tags
  When I choose P0 in the priority control and add the tag "checkout"
  Then the metadata panel reflects P0 and the checkout tag
  And the change is committed directly to the scenario's sidecar

Scenario: PUC-2 — Bulk tag/untag from the catalog (realizes BUC-2)
  Given I have multi-selected several scenarios in the catalog
  When I choose "Set priority → P1" from the bulk action bar
  Then every selected scenario shows P1
  And the bulk change is one commit

Scenario: PUC-3 — Filter the catalog by facet (realizes BUC-5, BUC-1)
  Given the left pane shows faceted filters generated from the collection schema
  When I check Priority P0 and tag "checkout"
  Then the catalog shows only scenarios matching both
  And each facet value shows its result count

Scenario: PUC-4 — A collection declares its fields (realizes BUC-4)
  Given a collection owner edits .collection.yaml to add a priority enum field
  When the collection is re-ingested
  Then the priority filter and the priority form control appear automatically

Scenario: PUC-5 — Migrate a collection to clean docs (product-only; enables BUC-3)
  Given a collection whose docs still carry top-of-doc frontmatter
  When the operator runs the frontmatter→sidecar migration
  Then each doc body becomes pure prose and a sidecar holds its metadata
  And rfc-app reads the collection identically before and after

11. UX Layout

11.1 Screen: Catalog (left pane) (serves PUC-3)

• Purpose: browse and filter a collection's entries.
• Layout (top → bottom):
  • Search: full-text box (existing).
  • Faceted filter groups (one per schema field + state): each is a collapsible group showing per-value result counts and multi-select checkboxes; tags-type fields include a "filter values…" search box to stay usable at 30+ values. (Chosen layout: faceted groups with counts — validated in brainstorming over flat chips.)
• States: happy: facets with counts · empty: "no entries match these filters" with a clear-filters action · loading: skeleton facets · error: "couldn't load facets" with retry.

11.2 Screen: Scenario detail — metadata panel (serves PUC-1)

• Purpose: view/edit one entry's metadata.
• Layout: a panel rendering one control per schema field — enum → single-select; tags → removable chips + add-tag input (with existing AI suggest where applicable); text → text input. The document body renders below as pure prose; metadata never appears inline in the body.
• States: read (role without edit) shows values, no controls · edit (authorized) shows controls · saving: inline spinner · error: field-level validation message (e.g. "P5 is not an allowed priority").

11.3 Screen: Catalog — bulk action bar (serves PUC-2)

• Purpose: apply a field value to many entries at once.
• Layout: selecting ≥1 row reveals a sticky action bar: "N selected · Set priority ▾ · Add tag ▾ · Remove tag ▾ · Clear". Each action targets one field; applying commits once.
• States: none selected: bar hidden · applying: bar shows progress · partial failure: toast naming entries that failed validation, others applied.

12. Technical Design

12.1 Invariants

• INV-1: The sidecar (<slug>.meta.yaml) is the source of truth for entry metadata; cached_rfcs is a derived index, fully rebuildable from git.
• INV-2: A document body (.md) never contains rfc-app metadata once migrated; metadata lives only in the sidecar.
• INV-3: Reading a collection never hard-fails on bad metadata — an invalid value against the schema surfaces as a warning and the entry still loads.
• INV-4: Metadata writes are authorized by scope-role (contributor+ on the collection); content-body edits keep their existing PR-review path.
• INV-5: A collection with no fields: block behaves exactly as today (free-form tags only) — the feature is additive and opt-in per collection.
• INV-6: Dual-read holds throughout: parser reads the sidecar if present, else legacy top-of-doc frontmatter, with identical resulting in-memory records.

12.2 High-level architecture

flowchart LR
  subgraph Git[content repo]
    CY[.collection.yaml<br/>fields: schema]
    MD[slug.md<br/>prose body]
    SC[slug.meta.yaml<br/>values]
  end
  CY --> ING[ingest / parser<br/>validate vs schema]
  MD --> ING
  SC --> ING
  ING --> DB[(cached_rfcs<br/>values + facet counts)]
  DB --> API[API: schema · list+filter · facets · edit]
  API --> FILT[left-pane faceted filters]
  API --> PANEL[detail metadata panel]
  API --> BULK[bulk select bar]
  PANEL -->|direct commit| SC
  BULK -->|1 commit| SC
  SC -.read from git.-> CONS[downstream consumers]

• ingest/parser — owns reading .collection.yaml schema + sidecars (or legacy frontmatter), validating values, and rebuilding cached_rfcs; must never treat the DB as authoritative.
• API — owns serving the schema, filtered lists with facet counts, and metadata edits; must never write metadata anywhere but the sidecar in git.

12.3 Data model & ownership

Entity	Owned by	Key fields	System of record
Collection field schema	collection owner	fields: {name → {type, values?, label}} in .collection.yaml	git
Entry metadata values	contributor	sidecar <slug>.meta.yaml: lifecycle (slug,title,state,owners,…) + schema fields (priority,tags,…)	git (sidecar)
Derived index	ingest	per-entry values + facet aggregations	cached_rfcs (SQLite, derived)

Field types (v1): enum (single-select; controlled by required values:), tags (multi-value; free-form unless values: given), text (free string). Unknown types are ignored with a warning (forward-compat).

Sidecar example:

slug: 01-01-0001-view-today-s-key-performance-metrics-at-a-glance
title: View today's key performance metrics at a glance
state: active
owners: [ben.stull]
priority: P1
tags: [dashboard, analytics]
owner: hasan

12.4 Interfaces & contracts

• GET /api/projects/<p>/collections/<c> — out: collection incl. fields schema. Errors: 404.
• GET /api/projects/<p>/collections/<c>/rfcs — in: filter params (?priority=P0&tags=checkout&state=active, repeatable for multi-value/OR-within-field, AND across fields) · out: entries with metadata values + facets: {field → {value → count}}. Errors: 400 on unknown field.
• POST /api/projects/<p>/collections/<c>/rfcs/<slug>/meta — in: {field: value, …} · out: updated values · effect: validate vs schema → write sidecar → commit directly → re-ingest entry. Errors: 403 (role), 422 (invalid value).
• POST /api/projects/<p>/collections/<c>/meta/bulk — in: {slugs: […], op: set|add|remove, field, value} · out: {applied: […], rejected: [{slug, reason}]} · effect: validate → write N sidecars → one commit → re-ingest. Errors: 403, 422.

12.5 Per–Product-Use-Case design

PUC-2 — Bulk tag/untag

sequenceDiagram
  actor U as Contributor
  participant C as Catalog UI
  participant A as API
  participant G as Git
  participant D as cached_rfcs
  U->>C: select rows, "Set priority P1"
  C->>A: POST /meta/bulk {slugs, set, priority, P1}
  A->>A: authz (scope-role) + validate vs schema
  A->>G: write N sidecars, 1 commit
  A->>D: re-ingest affected entries
  A-->>C: {applied, rejected}
  C-->>U: rows show P1; toast on any rejected

• Implementation: reuse the edit-meta git write-through, extended to (a) target the sidecar rather than frontmatter and (b) batch N files into one commit. Honors INV-1/INV-4.

PUC-5 — Migration

• Implementation: a tool (framework CLI verb / tools/ script) walks a collection, and for each entry with legacy frontmatter, writes <slug>.meta.yaml from the frontmatter and rewrites <slug>.md to the body only — one commit per collection, idempotent. Dual-read (INV-6) means this can run anytime; lazy migration converts stragglers on their first metadata edit.

12.6 Non-functional requirements & cross-cutting concerns

• Security & privacy: metadata edits gated by auth.effective_scope_role (contributor+ on the collection); no secrets in sidecars; git history records authorship.
• Performance & scale: facet counts computed from the derived DB; must stay responsive at ~1.2k entries with dozens of tag values (indexed value columns / aggregation query).
• Availability & resilience: bad metadata never blocks read (INV-3); a failed re-ingest leaves git authoritative and is recoverable by full rebuild.
• Observability: log each metadata commit (actor, field, entry count); warn-log schema validation failures encountered on ingest.
• Accessibility: facet groups and form controls keyboard-navigable; checkboxes labelled with value + count.

12.7 Key decisions & alternatives considered

Decision	Chosen	Alternatives	Why
Release modeling	Metadata only; releases downstream	First-class release entity in rfc-app; release-typed tags with behavior	Operator pulled ordering/ship-status out of rfc-app; metadata is the only needed primitive
Tag system shape	One generic typed-field system (Approach A)	Releases first-class + simple tags; namespaced facets	Tags/priority/custom are all just fields; one mechanism
Metadata storage	Sidecar per entry	Top-of-doc frontmatter (today); end-of-doc block; collection index file; DB-only	Clean docs + git-visible to consumers + locality per scenario
Left-pane filtering	Faceted groups with counts	Flat facet chips	Scales to the ~1.2k-scenario, many-tag ecomm corpus
Edit governance	Direct commit for authorized roles (bulk = 1 commit)	PR per change	Bulk planning is impractical via PR-per-toggle
Mgmt UI	Deferred; edit .collection.yaml in git	In-app field/vocab management in v1	Smallest coherent v1

12.8 Testing strategy

Unit tests for: schema parsing (fields: block, all types, missing block); sidecar read/write round-trip; dual-read equivalence (frontmatter vs sidecar produce identical records); schema validation (reject bad enum, accept free-form tag); facet aggregation; bulk op (set/add/remove, single commit, partial-rejection). Two-tier local-Docker→PPE for the API + git write-through. "Tested" = the PUC acceptance scenarios pass plus the migration is proven idempotent and reversible-on-read.

12.9 Failure modes, rollback & flags

• Failure mode: invalid value committed out-of-band → on ingest, warn + load entry with the raw value flagged (INV-3); not surfaced as a filter facet count error.
• Failure mode: re-ingest fails after commit → git is authoritative; full rebuild recovers.
• Migration rollback: dual-read means an un-migrated or partially-migrated corpus still works; the migration commit is revertible.
• Feature flag: the feature is inherently opt-in per collection (INV-5) — no global flag needed; absent a fields: block, behavior is unchanged.

13. Delivery Plan

13.1 Approach / strategy

Build the storage/compat foundation first (sidecars + dual-read + migration) so nothing breaks, then the schema, then read (filtering), then write (single, bulk). Each slice is shippable and non-breaking.

13.2 Slicing plan

SLICE-1 — Sidecar storage + dual-read + migration → completes PUC-5

• Depends on: —
• Definition of done: parser reads sidecar-if-present else legacy frontmatter (INV-6); migration tool splits frontmatter→sidecar idempotently; existing collections load byte-identically; tests green.

SLICE-2 — Collection field schema + validation → completes PUC-4

• Depends on: SLICE-1
• Definition of done: .collection.yaml fields: parsed (enum/tags/text); values validated on read (warn) and on write (reject); schema served via the collection API; no-fields: collections unchanged (INV-5).

SLICE-3 — Faceted left-pane filtering (read) → completes PUC-3

• Depends on: SLICE-2
• Definition of done: list endpoint returns facet counts + honors filter params; left pane renders faceted groups with counts and tag-value search; filters compose (AND across fields).

SLICE-4 — Single-entry metadata edit → completes PUC-1

• Depends on: SLICE-2
• Definition of done: detail metadata panel renders schema controls; POST …/meta validates, direct-commits the sidecar, re-ingests; authorized by scope-role (INV-4); lazy-migrates a legacy entry on first edit.

SLICE-5 — Bulk tag/untag → completes PUC-2

• Depends on: SLICE-3, SLICE-4
• Definition of done: multi-select + bulk action bar; POST …/meta/bulk applies set/add/remove as one commit; partial-rejection reported.

13.3 Rollout / launch plan

Pre-v1, single production: ship slices in order to the RFC deployment; each minor-version bump carries §20 changelog + upgrade steps. The opt-in-per-collection nature (INV-5) means a deployment adopts it only when it declares a fields: block and (optionally) runs the migration.

13.4 Risks & mitigations

Risk	L/I	Mitigation
Frontmatter→sidecar migration corrupts content	L/H	Dual-read; idempotent, revertible migration; body-byte-identity test
Doubling file count (sidecars) clutters corpus	M/L	Docs stay clean; sidecars are small/co-located; acceptable for one-file-per-scenario corpora
Direct-commit metadata edits bypass review	M/M	Scope-role gate (INV-4); content-body edits still PR'd; full git audit trail
Overlap/conflict with §22 S6 "type modules"	M/M	Position as §23, generalizing S6's per-type frontmatter; reconcile at the S6 SPEC merge
Facet aggregation slow at scale	L/M	Compute from indexed derived DB; measure at ~1.2k entries

14. Traceability matrix

Pain	Business UC	Product UC	Slice	Tests
PP-5	BUC-3	PUC-5	SLICE-1	test_dual_read_equiv, test_migration_idempotent
PP-7	BUC-4	PUC-4	SLICE-2	test_schema_parse, test_validation
PP-2	BUC-5, BUC-1	PUC-3	SLICE-3	test_facet_counts, test_filter_compose
PP-1, PP-3	BUC-1, BUC-4	PUC-1	SLICE-4	test_single_meta_commit, test_authz
PP-4	BUC-2	PUC-2	SLICE-5	test_bulk_one_commit, test_partial_reject
PP-6	BUC-3	(consumer reads git)	—	test_sidecar_schema_stable

15. Open Questions & Decisions log

Open

#	Question	Owner	Blocks
Q1	Do downstream consumers read sidecars from git, via rfc-app API, or both? (leaning git)	Ben	nothing v1
Q2	Should a multi-enum type (multi-select controlled) ship in v1 or later?	Ben	SLICE-2 scope
Q3	Exact §23 placement / reconciliation with §22 S6 type-modules	Ben	SPEC merge

Resolved

#	Decision	Resolution	Date
D1	Release behaviors (ordering, ship status, roadmap emit)	Out of rfc-app; downstream	2026-06-06
D2	Tag system shape	Approach A — one generic typed-field system	2026-06-06
D3	Metadata grain	Per entry (corpus already one file per scenario)	2026-06-06
D4	Schema location	.collection.yaml fields: block	2026-06-06
D5	Value storage	Sidecar <slug>.meta.yaml; doc body pure prose	2026-06-06
D6	Left-pane filtering	Faceted groups with counts	2026-06-06
D7	Edit governance	Direct commit for authorized roles; bulk = 1 commit	2026-06-06
D8	Management scope	Deferred; edit .collection.yaml in git for v1	2026-06-06

16. Glossary & References

• Sidecar — <slug>.meta.yaml, the per-entry metadata file that is the source of truth; keeps the .md body pure prose.
• Field schema — the fields: block in .collection.yaml declaring a collection's typed metadata fields.
• Facet — a schema field surfaced as a left-pane filter group with per-value counts.
• Downstream consumer — an external tool (e.g. a release planner) that reads corpus metadata from git; rfc-app does not model releases.
• References: retired BDD Release Planner (wiggleverse-ecomm-bdd-release-planner-app); §22 three-tier design (docs/design/2026-06-05-three-tier-projects-collections.md); SPEC §7.1 (left-pane filter), §9.5 (edit-meta), §20 (versioning), §22 Part B / S3 (scope-role).