chore(spec): merge p2-minutes-and-decisions-core-t3

.specter-prompt.txt

@@ -4,8 +4,8 @@ You are running HEADLESS. Do NOT use AskUserQuestion, interactive prompts, or sk
 Write files directly using the Write tool.
 
 ## Inputs
-- Change: p2-minutes-and-decisions-core-t2 (title: Minutes and Decisions — Core T2)
-- Context brief: openspec/changes/p2-minutes-and-decisions-core-t2/context-brief.md
+- Change: p2-minutes-and-decisions-core-t3 (title: Minutes and Decisions — Core T3)
+- Context brief: openspec/changes/p2-minutes-and-decisions-core-t3/context-brief.md
 - Data model ADR: openspec/architecture/adr-000-data-model.md
 
 Governance Domains:
@@ -29,13 +29,13 @@ Project guidelines: See ../project.md for workspace-wide standards
 
 ## Step 1: Read context
 Read these files COMPLETELY before writing anything:
-- openspec/changes/p2-minutes-and-decisions-core-t2/context-brief.md (features, stories, stakeholders, journeys)
+- openspec/changes/p2-minutes-and-decisions-core-t3/context-brief.md (features, stories, stakeholders, journeys)
 - openspec/architecture/adr-000-data-model.md (entity definitions)
 - All files in openspec/architecture/ (app ADRs)
 - All files in .claude/openspec/architecture/ (company ADRs, if present)
 
 ## Step 2: Get artifact templates
-Run: `openspec instructions proposal --change p2-minutes-and-decisions-core-t2 --json`
+Run: `openspec instructions proposal --change p2-minutes-and-decisions-core-t3 --json`
 This returns the template and rules for the proposal artifact.
 Then do the same for: design, specs, tasks (in that order).
 
@@ -54,8 +54,8 @@ For each artifact (proposal → design → specs → tasks):
 ## Step 4: Commit
 After all 4 artifacts are written:
 ```bash
-git add openspec/changes/p2-minutes-and-decisions-core-t2/
-git commit -m 'feat: Add OpenSpec change p2-minutes-and-decisions-core-t2 from Specter'
+git add openspec/changes/p2-minutes-and-decisions-core-t3/
+git commit -m 'feat: Add OpenSpec change p2-minutes-and-decisions-core-t3 from Specter'
 ```
 Do NOT push — the caller handles pushing.
 Do NOT create branches — stay on the current branch.

openspec/changes/p2-minutes-and-decisions-core-t3/.openspec.yaml

@@ -1,2 +1,2 @@
 schema: spec-driven
-created: 2026-04-17
+created: 2026-04-16

openspec/changes/p2-minutes-and-decisions-core-t3/context-brief.md

@@ -30,7 +30,7 @@ These specs were already decided/implemented. Use them as context.
 **Spec:** p1-schemas-and-data-model
 **Platform:** Nextcloud + OpenRegister
 
-## Features (6 total, spec-linked, sorted by market demand)
+## Features (6 total, sorted by market demand)
 
 ### Resolution Register
 **demand: 206** (66 tender mentions) | Category: core
@@ -101,10 +101,10 @@ Managing AV infrastructure in raadzaal and commissiekamers. Discussion systems,
 
 ### Handle Complex Multi-Domain Citizen Question
 A citizen has a question spanning multiple domains (e.g., housing benefit, parking permit, and social assistance). The front desk officer creates linked zaken or a combined intake and routes each to the correct department.
-**Trigger:** Citizen presents with multiple interconnected serv
+**Trigger:** Citizen presents with multiple interconnected service needs in 
 ... (truncated)
 
-## Features (16 total, spec-linked, sorted by market demand)
+## Features (16 total, sorted by market demand)
 
 ### Real-Time Action Item Tracking and Multi-Meeting Analytics
 **demand: 94** (21 tender mentions) | Category: core
@@ -825,16 +825,6 @@ All Conduction Nextcloud apps serve Dutch government users but must support mult
 - `l10n/en.json` is the identity-mapped source file (key == value).
 - Hardcoded Dutch strings in code MUST be converted to English keys with Dutch translations in `nl.json`.
 
-### Sentence Case for All UI Strings
-- All translation keys and user-facing strings MUST use **sentence case**: only the first word is capitalized.
-- Correct: `"Add directory"`, `"No results found"`, `"Delete selected"`, `"Save configuration"`
-- Wrong (title case): `"Add Directory"`, `"No Results Found"`, `"Delete Selected"`
-- Wrong (all lowercase): `"add directory"`, `"no results found"`
-- **Exceptions** that keep their capitalization:
-  - Proper nouns and product names: `"OpenRegister"`, `"Nextcloud"`, `"GitHub"`, `"DocuDesk"`
-  - Acronyms: `"API"`, `"URL"`, `"PDF"`, `"SOLR"`, `"JSON"`, `"RBAC"`, `"OAS"`
-  - Single-word strings still start with a capital: `"Delete"`, `"Search"`, `"Save"`
-
 ### Required Languages
 - Minimum: English (en) + Dutch (nl) translations.
 - `l10n/en.json` and `l10n/nl.json` MUST exist in every app with a UI.
@@ -855,16 +845,9 @@ All Conduction Nextcloud apps serve Dutch government users but must support mult
 - Date/number formatting: respect user locale via Nextcloud core.
 - Each app with OpenRegister: define `register-i18n` spec listing translatable fields.
 
-### Shared Component Library (@conduction/nextcloud-vue)
-- The shared library does NOT translate internally — it accepts pre-translated strings via props.
-- Components have English defaults for all label/text props (e.g., `addLabel="Add"`, `cancelLabel="Cancel"`).
-- Consumer apps are responsible for passing `t()` results as prop values.
-- The library lists `@nextcloud/l10n` as a peer dependency, not a direct dependency.
-
 ## Consequences
 - All apps maintain two translation files that must stay in sync.
 - Dutch strings used as translation keys (e.g., `t('app', 'Besluiten')`) are a violation — the English equivalent must be the key.
-- Title case in translation keys (e.g., `"Add Directory"`) is a violation — use sentence case (`"Add directory"`).
 - New features must include both `en.json` and `nl.json` entries before merging.
 
 ### ADR-008-testing
@@ -946,34 +929,18 @@ We want to unify these into a **single priority-scheduled container pool** so th
 
 ### Container types (priority order)
 
-| Priority | Type | Source | Container image | Model | Fallback |
-|----------|------|--------|-----------------|-------|----------|
-| 1 | **bugfix** | Hydra: fix iteration after review failure | `hydra-builder` | haiku | — |
-| 2 | **code-review** | Hydra: PR code review | `hydra-reviewer` | sonnet | opus |
-| 3 | **security-review** | Hydra: PR security review | `hydra-security` | sonnet | opus |
-| 4 | **build** | Hydra: initial spec build | `hydra-builder` | haiku | — |
-| 5 | **audit** | Hydra: codebase audit | `hydra-builder` | sonnet | opus |
-| 6 | **spec-generation** | Specter: push_spec_pipeline | `specter-llm-worker` | sonnet | haiku |
-| 7 | **schema-synthesis** | Specter: generate/dedup schemas | `specter-llm-worker` | haiku | — |
-| 8 | **classification** | Specter: classify/redistribute features | `specter-llm-worker` | haiku | — |
-| 9 | **translation** | Specter: translate requirements | `specter-llm-worker` | haiku | — |
-| 10 | **discovery** | Specter: research, feature extraction | `specter-llm-worker` | haiku | — |
-
-### Model strategy
-
-**Principle:** Use the cheapest model that can do the job. Reserve expensive models for judgment work.
-
-| Work type | Model | Rationale |
-|-----------|-------|-----------|
-| Build (implementation) | **Haiku** | Clear instructions (tasks.md, design.md). Pattern-following, not judgment. Faster and cheaper — 5 parallel Haiku builds burn far less quota than Sonnet. |
-| Fix (quality/browser/review) | **Haiku** | "Fix this PHPCS error" or "fix this review finding" — explicit, targeted corrections. |
-| Code review | **Sonnet → Opus** | Judgment work: spotting architectural issues, missed edge cases, style problems. Falls back to deeper model (Opus) when Sonnet quota is exhausted. |
-| Security review | **Sonnet → Opus** | Critical: injection vectors, auth bypasses, secret leaks. Same fallback logic. |
-| Audit | **Sonnet → Opus** | Full codebase analysis — needs depth. |
-
-**Quota optimization:** Claude Max plans have separate "Sonnet only" and "all models" weekly limits. By defaulting builders to Haiku, the Sonnet quota is reserved for reviews only (~20 turns each, 2 per PR). When Sonnet runs out, reviews fall back to the **deeper** model (Opus), not the shallower one — because reviews are the last line of defense before human approval.
-
-**Overrides:** Set `HYDRA_BUILDER_MODEL`, `HYDRA_REVIEWER_MODEL`, or `HYDRA_REVIEWER_FALLBACK_MODEL` env vars to change defaults.
+| Priority | Type | Source | Container image | Model |
+|----------|------|--------|-----------------|-------|
+| 1 | **bugfix** | Hydra: fix iteration after review failure | `hydra-builder` | sonnet |
+| 2 | **code-review** | Hydra: PR code review | `hydra-reviewer` | sonnet |
+| 3 | **security-review** | Hydra: PR security review | `hydra-security` | sonnet |
+| 4 | **build** | Hydra: initial spec build | `hydra-builder` | sonnet |
+| 5 | **audit** | Hydra: codebase audit | `hydra-builder` | sonnet |
+| 6 | **spec-generation** | Specter: push_spec_pipeline | `specter-llm-worker` | sonnet |
+| 7 | **schema-synthesis** | Specter: generate/dedup schemas | `specter-llm-worker` | haiku |
+| 8 | **classification** | Specter: classify/redistribute features | `specter-llm-worker` | haiku |
+| 9 | **translation** | Specter: translate requirements | `specter-llm-worker` | haiku |
+| 10 | **discovery** | Specter: research, feature extraction | `specter-llm-worker` | haiku |
 
 ### Architecture