skillpacks 0.1.1 → 0.1.2

package/packs/product-design/claude/ux-variations/archive/v0.18/SKILL.md

@@ -0,0 +1,176 @@
+---
+name: ux-variations
+description: Interview and plan multiple UX and UI variations for a product, page, or flow, including onboarding, typical workflows, sharing, collaboration, return use, and interface alternatives users can compare before locking a direction — and concrete visual/layout UI variations (component choices, spatial arrangements, information density)
+type: planning
+version: v0.18
+argument-hint: "[optional: app, page, flow, feature, or existing UI spec]"
+visual_tier: prototype
+---
+
+# UX Variations
+
+Invoke as `/ux-variations`.
+
+Use this skill when the user wants to explore multiple UX/UI directions before committing to a final experience. In the default product-design tree, this skill expands one specific user flow from `/user-flow-map` into alternate progression branches: different ways users can enter, advance, recover, hand off, complete, or abandon the flow. It then creates variation plans for flow progression, layout implications, navigation models, interaction patterns, component choices, content density, visual tone, and behavior so the user can compare, test, and decide which branch should move into `/ui-interview`.
+
+In the normal AFPS product route, use `/user-flow-map` first to establish the wireframe-tree root and name the user-flow branches. Then run `/ux-variations [specific-user-flow]` to explore alternate ways users can progress through that selected flow. After a variation branch is ready for UI treatment, route that branch to `/ui-interview [specific-ux-variation]` for visual mockup, alignment, approval/rejection, and next-branch routing.
+
+Use `/user-flow-map` first when the interface has no credible flow structure. Use this skill directly only when a user-flow map, current implementation, screenshot, prototype, explicit user prompt, or clear feature scope already identifies the flow being varied. Do not require a finalized UI requirements spec before default UX variation work; the point is to compare alternative progression paths before a single branch becomes a UI proposal.
+
+When invoked with `--layout-mode` (or when the user says "layout mode", "layout variations", or "UI variations"), this skill operates at the concrete component/layout level — it varies HOW the same content is presented visually, not WHAT the user flow is. Layout-mode is an explicit bounded mode for cases where both the flow and content contract are already fixed; otherwise default to progression-path UX variations. Layout-mode takes a fixed flow contract from `specs/user-flow-[topic].md` plus a fixed content contract from `specs/ui-requirements-[topic].md` or equivalent and generates 2-5 concrete visual/spatial approaches: different container patterns, detail views, navigation styles, density levels, and responsive strategies. Each variation is specified well enough to build as a lightweight implementation, then evaluated through `/uat --variant-evaluation` (check `.agents/project.json.enabled_packs` for `product-testing` — if `product-testing` is not enabled, recommend `npx skillpacks install product-testing` from the project shell, first) before `/consolidate-variations`.
+
+## Process
+
+### 0. Product-Path Scope Resolution
+
+Resolve research scope by product path before using code or app structure as a hint:
+
+1. If `$ARGUMENTS` names a non-archived `research/{slug}/` directory or a product-path ID whose `scope_path` points there, use that path. Treat `{slug}` as the product/app name, not the ICP, audience, or segment label.
+2. If `$ARGUMENTS` names only `research/_archive/{slug}/` or a manifest entry with `status: archived` or legacy `status: abandoned`, stop and warn that the path is archived; do not write or update scoped outputs there.
+3. Read `research/.progress.yaml` when present. Normalize legacy `active_path` to `active_paths` on read and write back `active_paths` on manifest updates. Treat legacy `abandoned` as `archived`; exclude `archived`, `abandoned`, `deferred`, `revisit_candidate`, `promoted`, and any `scope_path` under `research/_archive/` from active target selection.
+4. If active product paths exist in the manifest, use those paths. If multiple active paths exist, ask which one to target unless this skill explicitly supports cross-path output.
+5. If no active manifest target exists, list non-archived product directories under `research/`, excluding `research/_archive/` and dot directories. Auto-select only when exactly one exists; ask when multiple exist.
+6. If no product directories exist, use flat `research/` single-product mode.
+7. Detect monorepo/app/package structure only as a secondary hint. Suggest creating a missing `research/{slug}/` product path when code clearly exposes an app, but do not require code or monorepo detection before using `research/{slug}/`.
+
+When product path `{slug}` is active, read and write research under `research/{slug}/`, specs under `specs/{slug}/`, and treat top-level `research/*.md` files as flat-mode documents or cross-path summaries.
+
+1. **Resolve context**
+   - Read `.agents/project.json` if it exists.
+   - Read `README.md`, `AGENTS.md`, `CLAUDE.md`, relevant `docs/`, `specs/`, `research/`, task files, screenshots, route files, and component implementations when present.
+   - Read `research/.progress.yaml` when present. Normalize `active_path` (singular legacy) to `active_paths` (plural list) when reading; treat legacy `abandoned` as `archived` and exclude archived/deferred/revisit/promoted paths plus `research/_archive/` scopes from active target selection. Use `active_paths` as the product/app focuses and treat deferred `product_paths[]` as parked product directions, not required UX variants.
+   - Prefer existing `specs/user-flow-*.md` as the normal AFPS input and select one named user-flow branch as the variation surface.
+   - Also read `specs/ui-*.md`, product specs, journey maps, ICP research, positioning research, and user feedback as source evidence.
+   - If no credible flow structure exists, run or recommend `/user-flow-map` before developing variants.
+   - If the user explicitly requests layout-mode and no credible content/data/action contract exists, run or recommend `/ui-interview --requirements-only` before developing layout-mode variants.
+
+2. **Define the decision surface**
+   - Identify what the user is deciding: the selected user flow, onboarding, activation, typical workflow, sharing flow, collaboration model, purchase flow, editor, dashboard, settings, mobile experience, page layout, or another bounded surface.
+   - Identify the parent user-flow branch, the upstream `user-flow-map` source, and any sibling user flows or prior UX/UI proposals that this variation must coordinate with.
+   - Identify which dimensions may vary: first-run onboarding, activation, core workflow sequencing, sharing, invitations, permissions, collaboration, return-use loops, notifications, reminders, status surfaces, user or device handoffs, failure recovery, information architecture, navigation, page layout, task flow order, component model, data density, visual hierarchy, motion, copy tone, and mobile behavior.
+   - Identify fixed constraints: brand, stack, design system, must-keep components, accessibility, launch scope, performance, and business requirements.
+   - **Default progression-mode addition**: For each selected user flow, identify which parts may vary: entry route, sequencing, step granularity, user choice points, system automation, handoff points, review/edit loops, save-for-later behavior, cancellation, recovery, permissions, notifications, and completion criteria. Preserve the parent flow boundary while allowing progression branches to differ meaningfully.
+   - **Layout-mode addition**: In layout-mode, read `specs/user-flow-[topic].md` as the fixed screen-flow contract and `specs/ui-requirements-[topic].md` as the fixed content contract. The WHAT and flow order are locked; only the HOW varies. Layout dimensions that can vary:
+     - Container pattern: card grid, data table, list, kanban, timeline, tree, masonry
+     - Detail pattern: sidebar panel, full-page route, modal, drawer, inline expand, popover
+     - Navigation: top-nav, side-nav, tab-based, breadcrumb-driven, command-palette, hybrid
+     - Density: compact, comfortable, spacious
+     - Hierarchy: content-first, chrome-first, action-first
+     - Responsive strategy: reflow, collapse, separate mobile layout, progressive disclosure
+
+3. **Surface assumptions before probing**
+   - Present a UX Variation Assumptions Manifest before deep questioning.
+   - Tag assumptions with `[from spec]`, `[from codebase]`, `[from research]`, `[from artifact]`, or `[inferred]`.
+   - Cover target users, usage context, parent user flow, selected branch scope, first-run moment, activation event, "aha" threshold, typical repeat-use workflow, sharing and collaboration assumptions, permissions, return triggers, notifications, re-engagement assumptions, handoffs, pain points, locked versus open decisions including which parent-flow moments may vary, evaluation criteria, required variant breadth, prototype fidelity, implementation budget, success metrics, and selection method.
+   - Use AskUserQuestion to ask the user to confirm, correct, or flag assumptions before proceeding.
+   - Deliver every manifest/checklist/checkpoint the user must confirm inline as the final message text of its own turn; ask the confirmation question in the next turn. AskUserQuestion option previews may mirror the content as a supplement but are never the sole channel. Never emit it only as mid-turn text in a turn that ends with a tool call — harness rendering does not guarantee mid-turn text is shown. A confirmation question must never reference content the user has not been shown.
+
+4. **Interview for variation goals**
+   - Ask 1 to 3 focused questions per turn using AskUserQuestion.
+   - Default to maximally contrasting archetypes. Do not ask how different variants should be — assume dramatic contrast unless the user explicitly requests graduated steps.
+   - Default evaluation method is: build each variation, then run `/uat --variant-evaluation` (check `.agents/project.json.enabled_packs` for `product-testing` — if `product-testing` is not enabled, recommend `npx skillpacks install product-testing` from the project shell, first) so the user has task-based evidence before consolidation. Do not ask who will judge — assume solo evaluator building and gut-checking unless the user states otherwise.
+   - When presenting a design decision with 3+ plausible answers during the interview, always include "Make this a variant axis (test all approaches)" as an option. When the user has already chosen "test all" for a prior question in the same session, default subsequent ambiguous decisions to variant axes without asking.
+   - Establish which user flow from the wireframe tree is being expanded, what the variants must accomplish, which progression moments can vary (entry, sequencing, decision points, handoffs, recovery, completion, and re-entry), how a new user arrives and reaches first value, what the normal repeat workflow looks like, what users create/save/share/export/invite others into, what roles or permission levels exist, what notifications or status updates users expect, how users resume work after time away, what happens when a workflow is abandoned or blocked, which current interface parts work, which parts feel wrong or uncertain, what would make a variant unacceptable, and what evidence will decide the winner.
+   - When the user is unsure, recommend a practical default and explain why.
+   - **Layout-mode interview additions**: When in layout-mode, also ask:
+     - What is the primary user task on this page? (scan, search, create, compare, monitor, triage)
+     - How much data will typically be visible? (5 items, 50 items, 500 items)
+     - Are there reference apps or pages the user admires for this type of content?
+     - Are any layout patterns explicitly off the table?
+     - What is the build budget per variation? (quick prototype, medium fidelity, production-ready)
+     - What must the user do in each built variant before they are ready to consolidate?
+
+5. **Create distinct variation concepts**
+   - Produce 5 variations by default. Present the concepts for adjustment — do not ask the user to choose a count first.
+   - Each variation must be meaningfully different, not just a color or spacing change.
+   - At this stage, keep each concept lightweight: name, thesis, archetype, best-fit user/context, core workflow difference, major tradeoff, and rough complexity. Do not fully specify screens, controls, or implementation details yet.
+   - Useful archetypes include task-first workflow, data-dense operator console, guided step-by-step flow, onboarding-first activation path, collaboration-first workspace, sharing-first artifact flow, notification/status-driven workflow, role-based handoff workflow, visual canvas or board, command/search-first interface, mobile-first progressive disclosure, familiar SaaS dashboard, and editorial or showcase layout.
+   - Only choose archetypes that fit the product and user context.
+   - **Layout-mode archetypes** (use these instead of the UX-flow archetypes above when in layout-mode):
+     - Card grid: visual items in a responsive grid, good for browsing and scanning
+     - Data table: dense rows with sortable/filterable columns, good for operators and power users
+     - List + detail panel: master list on left, detail sidebar on right, good for email/messaging patterns
+     - Full-page detail: list view navigates to full-page item view, good for content-heavy items
+     - Kanban / board: columns representing states or categories, good for workflow and status tracking
+     - Timeline / feed: chronological stream, good for activity, logs, and social patterns
+     - Dashboard mosaic: mixed widget grid with charts, lists, and stats, good for monitoring and overview
+     - Split-pane workspace: resizable panels for parallel content, good for editors and comparison
+     - Command-first minimal: search/command bar with minimal chrome, good for keyboard-heavy power users
+     - Sidebar-driven: persistent sidebar navigation with content area, good for settings and multi-section apps
+
+6. **Concept selection checkpoint**
+   - Before fully specifying any variant, ask the user to adjust the concept set.
+   - Use bounded wording such as: "How should I adjust these UX variants before writing the final spec?"
+   - Present clear options:
+     - Keep all concepts
+     - Make one concept bolder or more extreme
+     - Add another concept
+   - Do not ask the user to remove or merge concepts before they have been built. Pre-build narrowing is consistently rejected.
+   - Ask the user to name the affected concept and briefly describe the change when they choose anything other than keeping all concepts.
+   - Recommend a practical default when evidence supports it; do not imply that variants have already been built or committed.
+   - Revise the concept set based on the answer before moving on.
+
+7. **Specify each approved variation enough to build**
+   - For each variation, define name and thesis, parent user flow and branch relationship, target user fit, onboarding and activation model, typical workflow sequence, progression model (how the user advances through the flow and how this differs from sibling variations), sharing and collaboration model, permissions model, return-use and notification model, failure recovery behavior, page and flow changes, navigation model, screen-by-screen layout, key components and controls, button and link behavior, spatial density, sizing, hierarchy, responsive behavior, visual tone, strengths, risks, failure modes, implementation complexity, prototype scope, and the user signal that would make this branch ready for `/ui-interview`.
+   - **Layout-mode variation spec additions**: In layout-mode, each variation spec must also include:
+     - Content-to-component mapping: which content requirement maps to which UI component
+     - Page regions with approximate proportions (e.g., sidebar 280px, content fluid, detail panel 400px)
+     - Primary content component (the main way items are displayed)
+     - Detail view pattern (how a single item's full details are accessed)
+     - Action placement (where create, edit, delete, and bulk actions live)
+     - Navigation pattern and placement
+     - Responsive behavior at 3 breakpoints (mobile ≤640px, tablet ≤1024px, desktop >1024px)
+     - Density approach (line-height, padding, font sizes, information per viewport)
+     - States rendering (how empty, loading, error, and partial states appear in this layout)
+     - Implementation file list (components, routes, layouts to create or modify)
+     - Estimated build time (hours)
+     - Variant evaluation task: the user task to perform in this variation before consolidation
+     - Evidence to capture: screenshots, notes, time-to-complete, friction points, and acceptance/rejection signals
+
+8. **Plan experimentation**
+   - Recommend serial full buildout of all approved variants. Do not recommend building a subset first — the user's consistent preference is to build all variants before evaluating.
+   - For prototype-stage product or feature work, prefer numerous small route-based experiments over one merged prototype when multiple workflows, layouts, densities, copy approaches, navigation models, or interaction patterns remain plausible. Name the route for each experiment, such as `/experiments/table-first`, `/experiments/command-first`, or the project's equivalent, and keep shared production infrastructure out of those routes unless explicitly approved.
+   - If route experiments imply materially different products, apps, ICPs, or product lines, update `research/.progress.yaml` with experiment product-path entries instead of making every divergent path a required UX variation. Include `id`, `label`, `source_skill: ux-variations`, `scope_path`, `status`, `reason`, `archive_reason`, `archived_at`, `promoted_at`, `evidence_refs`, `revisit_trigger`, `next_skill`, `pipeline_stage: ux-variations`, and `last_touched`.
+   - After progression variants are specified, route the next selected branch to `/ui-interview [specific-ux-variation]` for visual mockup and approval/rejection. If variants are built before UI interview, recommend `/uat --variant-evaluation` (check `.agents/project.json.enabled_packs` for `product-testing` — if `product-testing` is not enabled, recommend `npx skillpacks install product-testing` from the project shell, first) before `/consolidate-variations`. Consolidation is premature until evaluation evidence exists or the user explicitly says they reviewed the variants and is ready to converge.
+   - Define comparison criteria before selecting a winner.
+   - Include a lock-in checklist so the chosen direction becomes a decision record, not a vague preference.
+   - Include a UAT handoff checklist: target task for each variant, success criteria, side-by-side comparison questions, evidence to capture, tradeoffs to notice, and readiness criteria for `/consolidate-variations`.
+
+9. **Coverage checkpoint**
+   - Before concluding, summarize the variants, decision criteria, and experiment plan inline as the final message text of its own turn.
+   - In the next turn, use AskUserQuestion to ask whether any decision criteria, risks, validation steps, or implementation constraints are missing before writing deliverables.
+
+## Deliverables
+
+- Write the variation plan to `specs/ux-variations-[topic].md`.
+- Write the interview log to `ux-variations-[topic]-interview.md`.
+- Update `research/.progress.yaml` only when variant or route experiments create materially different product paths; downstream research remains active-path-only until a path is activated.
+- Include a branch-routing section that names the parent user flow, each UX variation branch, sibling flow/variation dependencies, and the recommended next `/ui-interview [specific-ux-variation]` branch.
+
+### Alignment Page
+
+When this skill produces durable deliverables (research, specs, plans, reports, prototypes, or any document output), build a full-depth HTML alignment page following `ALIGNMENT-PAGE.md` in this skill's directory. Output: `alignment/ux-variations-{topic}.html`.
+
+## Constraints
+
+- Do not present superficial variants that differ only by color palette, typography, or decorative treatment.
+- Do not treat default UX variation work as only visual layout exploration. Default variations should compare different ways users progress through a named user flow.
+- Do not choose a winner for the user unless the evidence clearly supports it and the user asked for a recommendation.
+- Do not defer all decisions to testing. State a recommended variant or experiment when evidence is sufficient.
+- Do not ignore implementation cost. A compelling variation still needs a prototype path and selection criteria.
+- Do not route directly from built UI variants to `/consolidate-variations`; insert `/uat --variant-evaluation` (check `.agents/project.json.enabled_packs` for `product-testing` — if `product-testing` is not enabled, recommend `npx skillpacks install product-testing` from the project shell, first) unless the user explicitly confirms they have already evaluated the variants.
+- Do not skip `/ui-interview` for a UX variation branch that needs a proposed UI, HTML visual mockup, or approval/rejection decision.
+- When recommending a skill from another pack, verify the pack is installed via `.agents/project.json` `enabled_packs`. If not installed, recommend `npx skillpacks install <pack-name>` from the project shell, before the target skill.
+- Do not enforce shared design constraints across variations. Each variation independently decides layout, density, color, navigation, and component choices. Only technical stack (framework, renderer, design system tokens) is shared unless the user explicitly locks a shared constraint.
+
+## Archive-First Replacement Policy
+
+- Before replacing or substantively rewriting an existing canonical research/spec document (`research/**/*.md`, `specs/**/*.md`, or `docs/specifications/**/*.md`), copy the current file to `docs/history/archive/YYYY-MM-DD/HHMMSS/<original-relative-path>`.
+- Preserve the archived snapshot exactly as it existed before the change; do not edit the archived copy after creating it.
+- After the archive snapshot exists, write the updated document to the original canonical path.
+- Report both the archive path and the updated canonical path in the final output.
+- New files do not need archive snapshots. Append-only updates do not need archive snapshots unless an existing section is regenerated or rewritten.
+
+## Default Shipping Contract
+
+Follow the shared shipping contract convention in CLAUDE.md.

package/packs/product-design/codex/brainstorm/ALIGNMENT-PAGE.md

@@ -1,6 +1,6 @@
 # Alignment Page — brainstorm
 
-When this skill produces durable deliverables (research, specs, plans, reports, prototypes, or any document output), build a full-depth HTML alignment page at `alignment/brainstorm-{topic}.html`. Use a normalized topic slug derived from the app, feature, research subject, report subject, or output filename.
+Alignment pages are optional for this operational skill. By default, report the outcome inline and write only the skill's normal durable artifacts (`tasks/*.md`, reports, queues, benchmark notes, status docs, or other skill-specific files). Build `alignment/brainstorm-{topic}.html` only when the user explicitly requests an alignment page or when the agent explicitly identifies a concrete clarification/review need that cannot be resolved cleanly in conversation or the normal artifacts. When an alignment page is created, use the full contract below and a normalized topic slug derived from the app, feature, research subject, report subject, or output filename.
 
 **Visual rendering tier.** Each alignment page declares a rendering tier:
 - `document` (default) — text, tables, approval gates. No charts or interactive elements.
@@ -9,6 +9,8 @@ When this skill produces durable deliverables (research, specs, plans, reports,
 
 The tier is set per skill via `visual_tier` in SKILL.md frontmatter (or detected by the upgrade script). The bundled ALIGNMENT-PAGE.md receives tier-appropriate rendering guidance from the generator.
 
+**Context intake.** This skill declares `context_intake: scoped` (Scoped intake). Ask only the focused questions needed to scope the research or analysis, then validate critical findings before building review artifacts.
+
 **Alignment lifecycle.** Alignment pages have three lifecycle states. `review` is the draft/pre-approval page: it contains the complete deliverable preview, evidence, required inline gate questions, section feedback controls, local section-feedback YAML only after a section feedback choice is selected, and one bottom `Compile Responses` control. `confirmed` is the post-approval page after the agent has consumed final compiled response YAML, made any approved user-requested edits, and written or updated the approved canonical artifacts. A confirmed page is finished and current for the completed alignment cycle, but it remains amendable when later research changes or extends the conclusions. `amended` is a future revision of a confirmed page: archive the previous confirmed page first, mark what changed, preserve the prior approval record, and highlight the new evidence, decisions, or scope that changed the earlier conclusion.
 
 **Central alignment index.** When a research-producing skill writes or updates an HTML alignment page, also maintain the repository's central alignment index. Look for `alignment/index.html` at the repository root; if the target repo documents a different central alignment-index path, use that path and report it. If the index does not exist, create it as a self-contained HTML page that works from `file://`, links to every active `alignment/*.html` page except itself, and includes the newly written page. Every index entry must include the page's alignment date in a muted meta span after the link, formatted `YYYY-MM-DD`, for example `<a href="page.html">Title</a> <span class="meta">2026-06-09</span>`. The alignment date is the review or confirmed date written inside the page itself; new pages use today's date. If the index exists, update it instead of blindly appending: preserve valid existing entries, metadata, and entry dates; remove duplicate entries for the same page path; keep links repo-relative; and include the new or updated page. When updating an existing index, add a date only to entries that are missing one, deriving it from the page's internal alignment date, or from file history when the page has no internal date. If a `new` marker is used, combine it with the date as `<span class="meta">new · YYYY-MM-DD</span>`, and remove stale `new` markers from entries that are no longer the latest addition.
@@ -57,7 +59,9 @@ The tier is set per skill via `visual_tier` in SKILL.md frontmatter (or detected
 
 **Report-only research gates.** For report-only or pre-approval research skills, the first `review` alignment page must explicitly contain proposed research scope, source plan, assumptions/confidence, recommended output path, proposed file changes, and approval gates before synthesized research begins and before any working packet, canonical research, spec, or task file is created or updated.
 
-**Staged research workflow.** Heavy report-first research uses a scope-first three-stage approval workflow. Stage 1 is minimal scope discovery only: inspect enough repository, user, and source context to propose research scope, source plan, assumptions, output paths, and approval questions. Build the `review` alignment page before synthesized research. The page must render the proposed scope, available source categories, known context, assumptions/confidence, proposed working-packet and canonical output paths, and research-scope approval gates. Stop for final compiled response YAML approval of the research scope. Do not synthesize findings, rank candidates, recommend a path, or create working packets, canonical research, spec, or task files in Stage 1. Stage 2 starts only after final compiled response YAML approves the research scope with no unresolved `needs-clarification`, unresolved `down` feedback, or other unresolved negative feedback. Perform the synthesized research, run required source/code checks, and write only a non-canonical working packet. Flat mode uses `research/_working/preliminary-brainstorm-research.md`; product-path mode uses `research/{slug}/_working/preliminary-brainstorm-research.md`. Existing raw evidence or search-log outputs may remain as supporting evidence when the skill contract already requires them, but synthesized deliverables stay in the working packet. Update the same `review` alignment page so it renders the full preliminary packet, evidence matrix, assumptions/confidence register, source coverage gaps, proposed canonical file changes, and artifact approval gates. Partial compiled responses revise the working packet and review page, then remain in Stage 2. Stage 3 consumes final compiled response YAML for artifact approval only when it has `approval_status: ready-for-agent-review` and no unresolved `needs-clarification`, unresolved `down` feedback, or other unresolved negative feedback. Apply approved edits first, archive the working packet to `docs/history/archive/YYYY-MM-DD/HHMMSS/<original-working-path>`, remove the active working packet, write the approved canonical artifacts to the unchanged output paths, and convert the page to `confirmed` with the approval record preserved.
+**Staged research workflow.** Heavy report-first research uses a scope-first three-stage approval workflow. Stage 1 is minimal scope discovery only: inspect enough repository, user, and source context to propose research scope, source plan, assumptions, output paths, and approval questions. Build the `review` alignment page before synthesized research. The page must render the proposed scope, available source categories, known context, assumptions/confidence, proposed working-packet and canonical output paths, research-scope approval gates, and a `Stage 2 Preview / Expected Review Format` section that shows the future Stage 2 review shape before research starts. Stop for final compiled response YAML approval of the research scope. Do not synthesize findings, rank candidates, recommend a path, or create working packets, canonical research, spec, or task files in Stage 1. Stage 2 starts only after final compiled response YAML approves the research scope with no unresolved `needs-clarification`, unresolved `down` feedback, or other unresolved negative feedback. Perform the synthesized research, run required source/code checks, and write only a non-canonical working packet. Flat mode uses `research/_working/preliminary-brainstorm-research.md`; product-path mode uses `research/{slug}/_working/preliminary-brainstorm-research.md`. Existing raw evidence or search-log outputs may remain as supporting evidence when the skill contract already requires them, but synthesized deliverables stay in the working packet. Update the same `review` alignment page so it renders the complete working-packet substance as structured HTML review UI: purpose-built sections, tables, matrices, gates, cards, and tier-appropriate charts or diagrams that preserve every packet section, finding, caveat, and decision detail without summary loss. Raw Markdown packet text may appear only as a supplemental source view after the rendered review UI; do not make a `Full Preliminary Packet` or `Full Working Packet` raw Markdown dump, giant `<pre><code>` block, link-only view, or source-only view the primary review surface. Include the evidence matrix, assumptions/confidence register, source coverage gaps, proposed canonical file changes, and artifact approval gates. Partial compiled responses revise the working packet and review page, then remain in Stage 2. Stage 3 consumes final compiled response YAML for artifact approval only when it has `approval_status: ready-for-agent-review` and no unresolved `needs-clarification`, unresolved `down` feedback, or other unresolved negative feedback. Apply approved edits first, archive the working packet to `docs/history/archive/YYYY-MM-DD/HHMMSS/<original-working-path>`, remove the active working packet, write the approved canonical artifacts to the unchanged output paths, and convert the page to `confirmed` with the approval record preserved.
+
+**Stage 2 review-page template.** Stage 1 scope-review pages must include a `Stage 2 Preview / Expected Review Format` section that previews the exact section sequence, table shapes, evidence treatment, visual density, and approval gates the user will see after research; make it specific enough for the user to flag preferred content formats, grouping, labels, visuals, and evidence density before the agent does heavy research. After scope approval and research, the Stage 2 review page must include these sections as structured HTML, not as raw Markdown dumps: `Research Scope Approved` status block with approved scope, source plan, output paths, and unresolved caveats; `Executive Findings` or a skill-specific top-findings section with confidence labels; `Evidence Matrix` mapping claim, evidence/source/path, inference, confidence, assumptions, and decision impact; `Working Packet Review` with the complete working-packet substance rendered as purpose-built HTML sections, lists, tables, matrices, cards, and tier-appropriate charts or diagrams; `Alternatives / Rejected or Lower-Confidence Findings`; `Source Coverage & Gaps`; `Assumptions / Confidence Register`; `Proposed Canonical Artifacts & File Changes`; a required `User Format Preferences` gate asking whether the layout, tables, grouping, visuals, labels, and evidence density match expectations; and a required `Final Artifact Approval` gate for Stage 3 readiness. Raw Markdown packet text, search logs, or source notes may appear only as supplemental source views after the rendered review UI and never replace the Stage 2 review sections.
 
 **Prototype tier rendering.** This skill's alignment page uses the `prototype` tier. In addition to charts and diagrams, functional interactive components (clickable flows, live CSS previews, drag interactions) are permitted. Prototypes are ephemeral previews, not production code. Each interactive component must include a reset button that restores initial state. Interactive components must be self-contained within the HTML file with no external dependencies. Include all visual tier requirements (chart snippets, table fallbacks, aria-labels, toggles) plus: state isolation (no localStorage/sessionStorage persistence), clear "this is a preview" labeling, and graceful degradation when JavaScript is disabled.
 

package/packs/product-design/codex/brainstorm/CHANGELOG.md

@@ -1,5 +1,9 @@
 # Changelog
 
+## v0.3 - 2026-06-12
+
+- Made alignment pages optional by default: report inline and write the skill's normal durable artifacts unless the user requests an alignment page or the agent identifies a concrete clarification/review need.
+
 ## v0.2 - 2026-05-30
 
 - Added product-path scope resolution that prefers non-archived `research/{slug}/` paths and active manifest paths before code or monorepo hints.

package/packs/product-design/codex/brainstorm/SKILL.md

@@ -2,9 +2,9 @@
 name: brainstorm
 description: Evaluate the codebase and suggest ideas to explore with $feature-interview
 type: planning
-version: v0.2
+version: v0.3
 argument-hint: "[optional focus area]"
-interview_depth: light
+context_intake: scoped
 visual_tier: prototype
 ---
 
@@ -78,7 +78,7 @@ Group suggestions by effort level (hours / days / weeks). Each suggestion should
 
 ## Alignment Page
 
-When this skill produces durable deliverables (research, specs, plans, reports, prototypes, or any document output), build a full-depth HTML alignment page following `ALIGNMENT-PAGE.md` in this skill's directory. Output: `alignment/brainstorm-{topic}.html`.
+By default, this skill reports results inline and writes only its normal durable artifacts (for example `tasks/*.md`, reports, queues, benchmark notes, status docs, or other skill-specific files). Do not build an alignment page automatically. Create `alignment/brainstorm-{topic}.html` only when the user explicitly requests an alignment page or when you explicitly identify a concrete clarification/review need that cannot be handled cleanly inline; when you create one, follow `ALIGNMENT-PAGE.md` in this skill's directory.
 
 ## Default Shipping Contract
 

package/packs/product-design/codex/brainstorm/archive/v0.2/SKILL.md

@@ -0,0 +1,85 @@
+---
+name: brainstorm
+description: Evaluate the codebase and suggest ideas to explore with $feature-interview
+type: planning
+version: v0.2
+argument-hint: "[optional focus area]"
+interview_depth: light
+visual_tier: prototype
+---
+
+# Brainstorm
+
+Invoke as `$brainstorm`.
+
+Evaluate the current codebase and generate actionable suggestions that the user can take into `$feature-interview` for human/agent alignment, planning-destination triage, and follow-up specification or roadmap work.
+
+## Process
+
+### 0. Product-Path Scope Resolution
+
+Resolve research scope by product path before using code or app structure as a hint:
+
+1. If `$ARGUMENTS` names a non-archived `research/{slug}/` directory or a product-path ID whose `scope_path` points there, use that path. Treat `{slug}` as the product/app name, not the ICP, audience, or segment label.
+2. If `$ARGUMENTS` names only `research/_archive/{slug}/` or a manifest entry with `status: archived` or legacy `status: abandoned`, stop and warn that the path is archived; do not write or update scoped outputs there.
+3. Read `research/.progress.yaml` when present. Normalize legacy `active_path` to `active_paths` on read and write back `active_paths` on manifest updates. Treat legacy `abandoned` as `archived`; exclude `archived`, `abandoned`, `deferred`, `revisit_candidate`, `promoted`, and any `scope_path` under `research/_archive/` from active target selection.
+4. If active product paths exist in the manifest, use those paths. If multiple active paths exist, ask which one to target unless this skill explicitly supports cross-path output.
+5. If no active manifest target exists, list non-archived product directories under `research/`, excluding `research/_archive/` and dot directories. Auto-select only when exactly one exists; ask when multiple exist.
+6. If no product directories exist, use flat `research/` single-product mode.
+7. Detect monorepo/app/package structure only as a secondary hint. Suggest creating a missing `research/{slug}/` product path when code clearly exposes an app, but do not require code or monorepo detection before using `research/{slug}/`.
+
+When product path `{slug}` is active, read and write research under `research/{slug}/`, specs under `specs/{slug}/`, and treat top-level `research/*.md` files as flat-mode documents or cross-path summaries.
+
+1. Read CLAUDE.md, README, package config, and key source files to understand the project.
+2. Check `tasks/roadmap.md`, `tasks/todo.md`, `tasks/manual-todo.md`, `tasks/record-todo.md`, `tasks/recurring-todo.md` (when they exist), and specs from `specs/` (or `spec.md`) if they exist — avoid suggesting things already planned or deferred as advisory records. Read `research/competitive-analysis.md` (or `research/{slug}/competitive-analysis.md`) if it exists — competitor gaps, market white space, and positioning weaknesses are high-signal inputs for ideation. Read `research/customer-feedback.md` (or `research/{slug}/customer-feedback.md`) if it exists — "Wrong" and "New" findings are highest-signal ideation input. Read `research/metrics.md` (or `research/{slug}/metrics.md`) if it exists — instrumentation gaps can generate ideas.
+3. Analyse the codebase across these dimensions:
+
+   **Strategic / Product**
+   - New features that would make the project significantly more useful or valuable
+   - New workflows or end-to-end automation the project could enable
+   - Product line expansion — adjacent use cases, new audiences, or complementary products the core could serve
+   - Integration opportunities with external tools, platforms, or APIs that multiply value
+
+   **Improvement**
+   - Missing capabilities the architecture is set up for
+   - Pain points, rough edges, or manual steps that could be automated
+   - Performance bottlenecks or low-hanging optimisations
+   - Developer experience friction
+
+   **Hygiene**
+   - Technical debt where code has outgrown its design
+   - Testing gaps in critical paths
+   - Security hardening opportunities
+
+   **Market Fit** (only when `research/icp.md` (or `research/{slug}/icp.md`), `research/mvp-gap.md` (or `research/{slug}/mvp-gap.md`), or `research/competitive-analysis.md` (or `research/{slug}/competitive-analysis.md`) exist)
+   - ICP alignment — features addressing ICP pain points that are missing or incomplete
+   - Journey gaps — steps where the product loses the user or customer
+   - Unaddressed MVP gaps from `research/mvp-gap.md` (or `research/{slug}/mvp-gap.md`) not yet in roadmap
+   - Competitive white space — features or capabilities no competitor offers well, from `research/competitive-analysis.md` (or `research/{slug}/competitive-analysis.md`) market gaps
+   - Competitor leapfrog — specific competitor weaknesses to exploit, or table-stakes features competitors have that you lack
+   - Positioning plays — ideas that sharpen differentiation against the competitive landscape
+4. If the user provides a focus area, scope the analysis there. Otherwise cover all dimensions.
+
+## Output
+
+Append the suggestions to `tasks/ideas.md` (do not overwrite existing content). Also display them to the user. When product-path scope is active, prefix each suggestion with the app name.
+
+Group suggestions by effort level (hours / days / weeks). Each suggestion should include:
+- A specific, actionable title
+- A one-line description with the concrete codebase signal that motivates it
+- A `$feature-interview <topic>` prompt the user can copy-paste
+
+## Constraints
+
+- Be specific and actionable — no vague aspirations.
+- Limit to 3–5 suggestions per effort level.
+- Do not suggest changes that conflict with CLAUDE.md conventions.
+- Do not repeat work already in `tasks/roadmap.md`, `tasks/todo.md`, `tasks/manual-todo.md`, `tasks/record-todo.md`, `tasks/recurring-todo.md`, or `specs/` (or `specs/{slug}/`).
+
+## Alignment Page
+
+When this skill produces durable deliverables (research, specs, plans, reports, prototypes, or any document output), build a full-depth HTML alignment page following `ALIGNMENT-PAGE.md` in this skill's directory. Output: `alignment/brainstorm-{topic}.html`.
+
+## Default Shipping Contract
+
+Follow the shared shipping contract convention in CLAUDE.md.

package/packs/product-design/codex/consolidate-variations/ALIGNMENT-PAGE.md

@@ -57,7 +57,9 @@ The tier is set per skill via `visual_tier` in SKILL.md frontmatter (or detected
 
 **Report-only research gates.** For report-only or pre-approval research skills, the first `review` alignment page must explicitly contain proposed research scope, source plan, assumptions/confidence, recommended output path, proposed file changes, and approval gates before synthesized research begins and before any working packet, canonical research, spec, or task file is created or updated.
 
-**Staged research workflow.** Heavy report-first research uses a scope-first three-stage approval workflow. Stage 1 is minimal scope discovery only: inspect enough repository, user, and source context to propose research scope, source plan, assumptions, output paths, and approval questions. Build the `review` alignment page before synthesized research. The page must render the proposed scope, available source categories, known context, assumptions/confidence, proposed working-packet and canonical output paths, and research-scope approval gates. Stop for final compiled response YAML approval of the research scope. Do not synthesize findings, rank candidates, recommend a path, or create working packets, canonical research, spec, or task files in Stage 1. Stage 2 starts only after final compiled response YAML approves the research scope with no unresolved `needs-clarification`, unresolved `down` feedback, or other unresolved negative feedback. Perform the synthesized research, run required source/code checks, and write only a non-canonical working packet. Flat mode uses `research/_working/preliminary-consolidate-variations-research.md`; product-path mode uses `research/{slug}/_working/preliminary-consolidate-variations-research.md`. Existing raw evidence or search-log outputs may remain as supporting evidence when the skill contract already requires them, but synthesized deliverables stay in the working packet. Update the same `review` alignment page so it renders the full preliminary packet, evidence matrix, assumptions/confidence register, source coverage gaps, proposed canonical file changes, and artifact approval gates. Partial compiled responses revise the working packet and review page, then remain in Stage 2. Stage 3 consumes final compiled response YAML for artifact approval only when it has `approval_status: ready-for-agent-review` and no unresolved `needs-clarification`, unresolved `down` feedback, or other unresolved negative feedback. Apply approved edits first, archive the working packet to `docs/history/archive/YYYY-MM-DD/HHMMSS/<original-working-path>`, remove the active working packet, write the approved canonical artifacts to the unchanged output paths, and convert the page to `confirmed` with the approval record preserved.
+**Staged research workflow.** Heavy report-first research uses a scope-first three-stage approval workflow. Stage 1 is minimal scope discovery only: inspect enough repository, user, and source context to propose research scope, source plan, assumptions, output paths, and approval questions. Build the `review` alignment page before synthesized research. The page must render the proposed scope, available source categories, known context, assumptions/confidence, proposed working-packet and canonical output paths, research-scope approval gates, and a `Stage 2 Preview / Expected Review Format` section that shows the future Stage 2 review shape before research starts. Stop for final compiled response YAML approval of the research scope. Do not synthesize findings, rank candidates, recommend a path, or create working packets, canonical research, spec, or task files in Stage 1. Stage 2 starts only after final compiled response YAML approves the research scope with no unresolved `needs-clarification`, unresolved `down` feedback, or other unresolved negative feedback. Perform the synthesized research, run required source/code checks, and write only a non-canonical working packet. Flat mode uses `research/_working/preliminary-consolidate-variations-research.md`; product-path mode uses `research/{slug}/_working/preliminary-consolidate-variations-research.md`. Existing raw evidence or search-log outputs may remain as supporting evidence when the skill contract already requires them, but synthesized deliverables stay in the working packet. Update the same `review` alignment page so it renders the complete working-packet substance as structured HTML review UI: purpose-built sections, tables, matrices, gates, cards, and tier-appropriate charts or diagrams that preserve every packet section, finding, caveat, and decision detail without summary loss. Raw Markdown packet text may appear only as a supplemental source view after the rendered review UI; do not make a `Full Preliminary Packet` or `Full Working Packet` raw Markdown dump, giant `<pre><code>` block, link-only view, or source-only view the primary review surface. Include the evidence matrix, assumptions/confidence register, source coverage gaps, proposed canonical file changes, and artifact approval gates. Partial compiled responses revise the working packet and review page, then remain in Stage 2. Stage 3 consumes final compiled response YAML for artifact approval only when it has `approval_status: ready-for-agent-review` and no unresolved `needs-clarification`, unresolved `down` feedback, or other unresolved negative feedback. Apply approved edits first, archive the working packet to `docs/history/archive/YYYY-MM-DD/HHMMSS/<original-working-path>`, remove the active working packet, write the approved canonical artifacts to the unchanged output paths, and convert the page to `confirmed` with the approval record preserved.
+
+**Stage 2 review-page template.** Stage 1 scope-review pages must include a `Stage 2 Preview / Expected Review Format` section that previews the exact section sequence, table shapes, evidence treatment, visual density, and approval gates the user will see after research; make it specific enough for the user to flag preferred content formats, grouping, labels, visuals, and evidence density before the agent does heavy research. After scope approval and research, the Stage 2 review page must include these sections as structured HTML, not as raw Markdown dumps: `Research Scope Approved` status block with approved scope, source plan, output paths, and unresolved caveats; `Executive Findings` or a skill-specific top-findings section with confidence labels; `Evidence Matrix` mapping claim, evidence/source/path, inference, confidence, assumptions, and decision impact; `Working Packet Review` with the complete working-packet substance rendered as purpose-built HTML sections, lists, tables, matrices, cards, and tier-appropriate charts or diagrams; `Alternatives / Rejected or Lower-Confidence Findings`; `Source Coverage & Gaps`; `Assumptions / Confidence Register`; `Proposed Canonical Artifacts & File Changes`; a required `User Format Preferences` gate asking whether the layout, tables, grouping, visuals, labels, and evidence density match expectations; and a required `Final Artifact Approval` gate for Stage 3 readiness. Raw Markdown packet text, search logs, or source notes may appear only as supplemental source views after the rendered review UI and never replace the Stage 2 review sections.
 
 **Consolidation-specific gates.** Render UAT evidence coverage, variation verdicts, selected concept, rejected alternatives, unresolved assumptions, output-location/change-scope decisions, and coverage checkpoint as gates. Apply the shared artifact-destination/proposed-file-changes de-duplication rule: combine them into one visual gate when they ask only the same path-destination question, and render separate gates when destination approval differs from downstream mutation scope, timing, or allowed files.
 

package/packs/product-design/codex/consolidate-variations/CHANGELOG.md

@@ -1,5 +1,14 @@
 # Changelog
 
+## v0.13 - 2026-06-13
+
+- Updated consolidation to consume `design/` variation plans, UI requirements, and flow-tree manifests instead of pre-prototype `specs/` files.
+- Clarified that consolidation outputs a consolidated prototype under `prototypes/{topic}/consolidated/` and records consolidation state in the design flow-tree manifest.
+
+## v0.12 - 2026-06-12
+
+- Standardized active pack and skill install guidance on `npx skillpacks install <pack-or-skill>` instead of agent-native `/pack install` or `$pack install` recommendations.
+
 ## v0.11 - 2026-06-11
 
 - Added npm-aware install-route wording for product-testing recommendations while preserving Codex `$pack install ...` routes.

package/packs/product-design/codex/consolidate-variations/SKILL.md

@@ -1,8 +1,8 @@
 ---
 name: consolidate-variations
-description: Compare multiple built UI variations after UAT evidence, interview the user on what works and what does not, resolve conflicts, and produce a final implementation-ready UI specification
+description: Compare multiple built UI variations after UAT evidence, interview the user on what works and what does not, resolve conflicts, and produce a final consolidated prototype
 type: planning
-version: v0.11
+version: v0.13
 argument-hint: "[optional: topic, page, or path to variation specs]"
 visual_tier: prototype
 ---
@@ -11,7 +11,7 @@ visual_tier: prototype
 
 Invoke as `$consolidate-variations`.
 
-Use this skill after the user has built and evaluated multiple UI layout variations (typically generated via `$ux-variations --layout-mode`, built via `$exec`, and evaluated via `$uat --variant-evaluation` (check `.agents/project.json.enabled_packs` for `product-testing` — if `product-testing` is not enabled, recommend `$pack install product-testing` inside Codex, or `npx skillpacks install product-testing` from the project shell, first)). This skill compares the variations, interviews the user on what works and what does not in each one, cherry-picks the best elements, resolves conflicts where preferred choices are incompatible, and produces a single consolidated implementation-ready UI specification.
+Use this skill after the user has built and evaluated multiple UI layout variations (typically generated via `$ux-variations --layout-mode`, built via `$prototype`, and evaluated via `$uat --variant-evaluation` (check `.agents/project.json.enabled_packs` for `product-testing` — if `product-testing` is not enabled, recommend `npx skillpacks install product-testing` from the project shell, first)). This skill compares the variations, interviews the user on what works and what does not in each one, cherry-picks the best elements, resolves conflicts where preferred choices are incompatible, and produces a single consolidated prototype for post-prototype production specification.
 
 Users with manually built variations can also use this skill directly, but consolidation should not happen before the user has reviewed the variants and captured evidence.
 
@@ -29,21 +29,22 @@ Resolve research scope by product path before using code or app structure as a h
 6. If no product directories exist, use flat `research/` single-product mode.
 7. Detect monorepo/app/package structure only as a secondary hint. Suggest creating a missing `research/{slug}/` product path when code clearly exposes an app, but do not require code or monorepo detection before using `research/{slug}/`.
 
-When product path `{slug}` is active, read and write research under `research/{slug}/`, specs under `specs/{slug}/`, and treat top-level `research/*.md` files as flat-mode documents or cross-path summaries.
+When product path `{slug}` is active, read research under `research/{slug}/`, read pre-prototype design artifacts under `design/{slug}/`, write prototype output under `prototypes/{topic}/`, and treat top-level `research/*.md` and `design/*.md` files as flat-mode documents or cross-path summaries.
 
 1. **Resolve context**
    - Read `.agents/project.json` if it exists.
    - Read `README.md`, `AGENTS.md`, `CLAUDE.md`, relevant `docs/`, `specs/`, `research/`, route files, component directories, and design artifacts when present.
-   - Locate the variation spec: `specs/ui-layout-variations-[topic].md` or `specs/ux-variations-[topic].md`.
-   - Locate the content requirements: `specs/ui-requirements-[topic].md` or equivalent content contract.
+   - Locate the variation plan: `design/ui-layout-variations-[topic].md` or `design/ux-variations-[topic].md`.
+   - Locate the content requirements: `design/ui-requirements-[topic].md` or equivalent content contract.
+   - Locate the flow-tree manifest: `design/flow-tree-[topic].yaml` or `design/{slug}/flow-tree-{topic}.yaml` when present.
    - Locate variant evaluation evidence: `research/uat-variant-evaluation-[topic].md`, `research/uat-plan.md` result logs, screenshots, notes, recordings, or explicit user-provided review notes.
    - Locate built implementations: scan route files, component directories, and any variation-specific directories or branches.
    - If the variation spec or implementations cannot be found, ask the user to point to them.
 
 2. **Evidence gate**
-   - If no evaluation evidence exists and the user has not explicitly said they already reviewed the variants and is ready to converge, stop and recommend `$uat --variant-evaluation` (check `.agents/project.json.enabled_packs` for `product-testing` — if `product-testing` is not enabled, recommend `$pack install product-testing` inside Codex, or `npx skillpacks install product-testing` from the project shell, first).
+   - If no evaluation evidence exists and the user has not explicitly said they already reviewed the variants and is ready to converge, stop and recommend `$uat --variant-evaluation` (check `.agents/project.json.enabled_packs` for `product-testing` — if `product-testing` is not enabled, recommend `npx skillpacks install product-testing` from the project shell, first).
    - Do not infer a winner from specs alone. Built variants need hands-on review or explicit user readiness before consolidation.
-   - If some variants are unreviewed, ask whether to exclude them, evaluate them first via `$uat --variant-evaluation` (check `.agents/project.json.enabled_packs` for `product-testing` — if `product-testing` is not enabled, recommend `$pack install product-testing` inside Codex, or `npx skillpacks install product-testing` from the project shell, first), or include them as spec-only references.
+   - If some variants are unreviewed, ask whether to exclude them, evaluate them first via `$uat --variant-evaluation` (check `.agents/project.json.enabled_packs` for `product-testing` — if `product-testing` is not enabled, recommend `npx skillpacks install product-testing` from the project shell, first), or include them as spec-only references.
 
 3. **Present variation inventory**
    - List each variation with a one-line summary of its approach.
@@ -90,7 +91,7 @@ When product path `{slug}` is active, read and write research under `research/{s
    - Ask the user to confirm the consolidated design before building the prototype.
 
 7. **Coverage checkpoint**
-   - Verify every content requirement from `specs/ui-requirements-[topic].md` has a UI home in the consolidated spec.
+   - Verify every content requirement from `design/ui-requirements-[topic].md` has a UI home in the consolidated prototype.
    - Verify every user action has a placement: button, menu item, keyboard shortcut, or gesture.
    - Verify all states are accounted for: empty, loading, error, partial, full, offline, permission-denied.
    - Flag any gaps and resolve them before writing.
@@ -98,7 +99,8 @@ When product path `{slug}` is active, read and write research under `research/{s
 ## Deliverables
 
 - Write the consolidated prototype to `prototypes/{topic}/consolidated/`.
-- Write the consolidation interview log to `consolidate-variations-[topic]-interview.md`.
+- Write the consolidation interview log to `design/consolidate-variations-[topic]-interview.md` in flat mode or `design/{slug}/consolidate-variations-[topic]-interview.md` in product-path mode.
+- Update the scoped flow-tree manifest to mark consolidated branches as `consolidated` or `promoted-to-prototype` when applicable.
 
 ### Alignment Page
 
@@ -109,13 +111,13 @@ When this skill produces durable deliverables (research, specs, plans, reports,
 - Do not proceed without evaluation evidence unless the user explicitly says they have reviewed the variants and is ready to converge.
 - Do not pick winners without user input. Present the matrix and let the user decide.
 - Do not ignore conflicts. If two preferred choices are spatially or functionally incompatible, surface the tension and resolve it explicitly.
-- The consolidated spec must be at least as detailed as a `$ui-interview` output — implementation-ready, not a summary.
+- The consolidated prototype must preserve the approved UI branch detail from `$ui-interview` and be concrete enough for `$spec-interview` to extract production implementation requirements.
 - Do not lose content requirements. Every data field, action, and state from the requirements spec must appear in the final design.
 - Do not bias toward the first or last variation reviewed. Present them neutrally and let the user's feedback and evaluation evidence drive the outcome.
 
 ## Archive-First Replacement Policy
 
-- Before replacing or substantively rewriting an existing canonical research/spec document (`research/**/*.md`, `specs/**/*.md`, or `docs/specifications/**/*.md`), copy the current file to `docs/history/archive/YYYY-MM-DD/HHMMSS/<original-relative-path>`.
+- Before replacing or substantively rewriting an existing canonical research/design/spec document (`research/**/*.md`, `design/**/*.md`, `specs/**/*.md`, or `docs/specifications/**/*.md`), copy the current file to `docs/history/archive/YYYY-MM-DD/HHMMSS/<original-relative-path>`.
 - Preserve the archived snapshot exactly as it existed before the change; do not edit the archived copy after creating it.
 - After the archive snapshot exists, write the updated document to the original canonical path.
 - Report both the archive path and the updated canonical path in the final output.