@keighleykodric/weeve 0.1.0 → 0.1.2

package/examples/gtm/demo/weeve/shared/roadmap.md

@@ -0,0 +1,19 @@
+# Roadmap — Relay
+
+_Ask Sarah (PM) for the full roadmap. This is what's public knowledge on the team._
+
+## Shipping soon
+
+- LinkedIn step analytics (open/reply rates per step, not just per sequence)
+- AI personalization improvements — more CRM field support
+
+## Planned
+
+- Free trial / PLG motion — actively debated, no decision made
+- Salesforce native app (currently an API integration)
+- Call step (integrated calling, not just call tasks)
+
+---
+
+_For details, see Notion > Product > Relay Roadmap._
+_The PLG debate has been ongoing for 6 weeks without a decision._

package/examples/gtm/demo/weeve/shared/weeve-guardrails.md

@@ -0,0 +1,24 @@
+# weeve-guardrails.md — Relay
+
+## Hard limits
+- Sales-led only for now — no self-serve or PLG until onboarding is figured out
+- Salesforce and HubSpot integrations are required for enterprise deals — cannot remove or degrade
+- No price reduction — $99/seat/month is the floor; discounting has already compressed margins
+
+## GTM constraints
+- 4-person sales team (1 AE, 2 SDRs, 1 CS); no marketing hire until Q3
+- Content/SEO is not a current motion — all pipeline is outbound + referral
+- ICP is defined as "B2B SaaS, 10–50 person sales team, Series A–B" — but it's informal, not validated
+- Average sales cycle is 18 days; anything over 30 days has a <10% close rate
+
+## Non-negotiables
+- "Quality over volume" positioning must be preserved — team believes it's core to the brand
+- No reseller or channel motion — direct only
+- Customer success must respond within 4 hours on business days (SLA in contracts)
+
+## Pipeline data (last 90 days)
+- 148 opportunities created, 31 closed-won, 21% win rate
+- Average deal size: $780/month (avg 7.9 seats)
+- Top loss reason (by count): "evaluating Apollo" (38%), "no budget right now" (29%), "went with Outreach" (18%)
+- 2 reps have 38–41% win rates; other 2 reps have 12–14% win rates
+- Churn: 18% annual; 60% of churned customers never completed setup of their first sequence

package/examples/gtm/demo/weeve/shared/weeve-intake.md

@@ -0,0 +1,18 @@
+# weeve-intake.md — Relay observations
+
+---
+
+## Captured observations
+
+- Win rate is 21%; two reps are at 38–41%, two are at 12–14% — nobody has analyzed why
+- Top loss reason is "evaluating Apollo" (38% of losses) — but Apollo is positioned as higher-volume/lower-quality, which should be a differentiator, not a loss reason
+- Website hero copy says "The smarter outbound platform" — doesn't mention "quality over volume" at all
+- Sales deck uses "personalization at scale" — the phrase Apollo also uses heavily
+- 60% of churned customers never set up a sequence in the first 14 days
+- CS doesn't have a structured onboarding playbook — every CSM does it differently
+- Three customers in the last month said "I didn't realize it could do LinkedIn steps too" — feature awareness gap
+- No formal ICP document exists; the 4 salespeople describe the ICP differently when asked
+- One rep (highest win rate) always leads with the AI personalization demo; others lead with the sequence builder
+- The case studies on the website are 14 months old; they don't mention the AI personalization feature
+- Pricing page has 3 tiers but rep feedback is "we only ever sell the middle tier; top tier is never bought"
+- Referrals account for 31% of new pipeline but there's no formal referral program

package/examples/gtm/demo/weeve/shared/weeve-project.md

@@ -0,0 +1,25 @@
+# Relay — project brief
+
+**What it is:** Relay is a B2B outbound sales automation tool. Sales reps build multi-step sequences (email, LinkedIn, call tasks), and Relay personalizes outreach using CRM data and prospect research. Positioned against Apollo and Outreach for teams that find those tools "too complex."
+
+**Stage:** Series A ($8M). 210 paying customers, mostly 10–50 person sales teams. ARR ~$2.4M. Grown from $800K to $2.4M in 14 months.
+
+**Core product:**
+- Sequence builder (email + LinkedIn + tasks, drag-and-drop)
+- AI personalization (pulls from LinkedIn, company news, CRM fields)
+- Inbox management (reply detection, auto-pause on response)
+- Analytics (open rates, reply rates, meeting booked rate by sequence)
+- CRM integrations: Salesforce, HubSpot, Pipedrive
+
+**The GTM problem:**
+Revenue is growing but it's noisy. Win rate is 21% (team thinks it should be 35%+). The messaging has evolved organically and is inconsistent across the website, the sales deck, and what reps actually say on calls. Two reps are consistently outperforming their peers — the team hasn't codified why. Churn is 18% annually, concentrated in customers who don't set up a sequence in the first 14 days.
+
+**Competitive situation:**
+- Apollo: lower-priced, data-heavy, perceived as spammy — Relay's main differentiator is "quality over volume"
+- Outreach/Salesloft: enterprise, complex, $50K+ contracts — Relay wins deals where budget is the objection
+- Instantly/Lemlist: email-only, founder-led sales teams — adjacent but different motion
+
+**Open questions:**
+- The "quality over volume" positioning resonates with some buyers but not others — is it the right message?
+- Why are two reps winning at 2× the rate of the rest? Is it repeatable?
+- Should we add a PLG motion (free trial) or stay sales-led?

package/examples/gtm/demo/weeve.yaml

@@ -0,0 +1,22 @@
+# weave.yaml — gtm preset demo project
+schema_version: "0.4"
+
+project:
+  name: Relay
+  tag: RLY
+  docs_root: ./weeve
+  depends_on:
+    - tag: BCN
+      name: Beacon
+      relationship: market-intelligence  # Relay's positioning and ICP are informed by Beacon's
+                                         # B2B analytics. Beacon's strategy findings flow into
+                                         # Relay's go-to-market decisions.
+
+roads:
+  - name: startup
+    path: roads/startup
+
+# Lanes in this preset: compass, maven, pitch, pilot
+# Demo project: Relay — outbound sales automation SaaS
+# Depends on: Beacon (strategy preset) — try running both to see cross-project signals in Pilot
+# Run /compass-intro to get started

package/examples/gtm/weeve.yaml

@@ -0,0 +1,14 @@
+# weeve.yaml — gtm preset
+# Generated by: weeve init gtm
+schema_version: "0.4"
+
+project:
+  name: My Project             # Replace with your project name
+  tag: MYP                     # Short ID — matches vault Projects/<tag>/ folder
+  # docs_root: ./weeve          # Uncomment to use a local path instead of vault
+
+roads:
+  - name: startup
+    path: roads/startup
+
+# Lanes in this preset: compass, maven, pitch, pilot

package/examples/product/CODEOWNERS.template

@@ -0,0 +1,10 @@
+# Weeve lane ownership
+# Copy to .github/CODEOWNERS and replace team handles with your org's teams
+# Requires GitHub branch protection with "Require review from Code Owners" enabled
+
+weeve/compass/       @leadership
+weeve/layers-plus/   @product
+weeve/ally/          @design
+weeve/rubric/        @design
+weeve/felt/          @product
+weeve/pilot/         @leadership

package/examples/product/README.md

@@ -0,0 +1,67 @@
+# product preset
+
+## Who this is for
+
+The `product` preset is for product teams building consumer or B2B products where UX quality, user research alignment, and accessibility are first-class concerns. It extends the `core` preset with `felt` — a lane that captures qualitative user sentiment and experience signals — giving product managers and designers a fuller picture before committing to a release. Use this preset when you have a dedicated product and design function and want their workflows formalized and review-gated.
+
+## What you get
+
+- **compass-signals.md** — prioritized strategic signals ranked by market urgency and opportunity size
+- **compass-context.md** — competitive and market position snapshot to anchor product decisions
+- **layers-plus-map.md** — audience segmentation with motivational layers and decision drivers
+- **ally-principles.md** — accessibility requirements and inclusive design constraints for the product
+- **rubric-scorecard.md** — UX heuristic scorecard with severity-ranked regressions
+- **felt-pulse.md** — qualitative user sentiment summary: themes, friction points, and moments of delight
+- **pilot-brief.md** — launch readiness brief combining strategic, UX, accessibility, and sentiment signals
+
+## Install
+
+```bash
+npx weeve add product
+```
+
+Equivalent individual lanes:
+
+```bash
+npx weeve add compass layers-plus ally rubric felt pilot
+```
+
+## First run sequence
+
+Run these skills in order. Steps marked **(parallel)** can be run simultaneously once their prerequisites are complete.
+
+1. **compass** — Scans market signals and competitive context to establish the strategic backdrop for product decisions.
+2. **layers-plus** — Maps audience segments into motivational layers; uses compass context to weight which segments are most relevant to the current release.
+3. **(parallel)* ally** — Audits the product for accessibility gaps and outputs binding design constraints.
+4. **(parallel)* rubric** — Scores UX flows against quality heuristics, calibrated against the layers-plus audience map.
+5. **(parallel)* felt** — Synthesizes qualitative user research, session notes, and feedback into a structured sentiment report.
+6. **pilot** — Combines all upstream signals into a launch readiness brief with a go/no-go recommendation.
+
+## What you'll have at the end
+
+| File | Description |
+|---|---|
+| `weeve/compass/compass-signals.md` | Ranked strategic signals with evidence and urgency scores |
+| `weeve/compass/compass-context.md` | Market and competitive position snapshot |
+| `weeve/layers-plus/layers-plus-map.md` | Audience layer map with needs, motivations, and decision drivers |
+| `weeve/ally/ally-principles.md` | Accessibility constraints and inclusive design requirements |
+| `weeve/rubric/rubric-scorecard.md` | UX heuristic scorecard with severity-ranked issues |
+| `weeve/felt/felt-pulse.md` | Qualitative user sentiment: themes, friction points, and delight moments |
+| `weeve/pilot/pilot-brief.md` | Launch readiness brief with go/no-go recommendation |
+
+## CODEOWNERS
+
+Copy to `.github/CODEOWNERS`:
+
+```
+weeve/compass/       @leadership
+weeve/layers-plus/   @product
+weeve/ally/          @design
+weeve/rubric/        @design
+weeve/felt/          @product
+weeve/pilot/         @leadership
+```
+
+## Time estimate
+
+First full run: **60–90 minutes**, with the bulk of time in `felt` if you have extensive user research artifacts to synthesize. Teams with lightweight research inputs can complete felt in under 15 minutes. Subsequent runs: 20–35 minutes.

package/examples/product/demo/README.md

@@ -0,0 +1,21 @@
+# Canvas — product preset demo
+
+A no-code form and workflow builder for SMBs. Use this demo to try the **product** preset (Compass, Layers+, Ally, Rubric, Felt, Pilot).
+
+## Run it
+
+```bash
+weeve add product
+/compass-intro      # strategic positioning
+/layers-plus-intro  # 7-layer product evaluation
+/felt-intro         # UX research — personas, journeys
+/ally-intro         # accessibility audit
+/rubric-intro       # design system health
+/pilot-priorities   # synthesize all lanes
+```
+
+Outputs land in `./weeve/<lane>/`.
+
+## The project
+
+Canvas has real product tension: an NPS of 42 with recurring qualitative complaints about the same feature, no user research in 18 months, a known iOS bug that's 4 months old, and enterprise interest the team isn't sure how to handle. Good material for every lane in the preset.

package/examples/product/demo/weeve/ally/ally-intake.md

@@ -0,0 +1,16 @@
+# ally-intake — Canvas
+# Source: browser devtools check, team observation
+
+_Feed to `/ally-intake`. Not a systematic audit — just what's been noticed._
+
+---
+
+- Placeholder text in the default theme fails WCAG AA contrast. Checked with browser devtools.
+- Embedded iframe forms: never tested with a screen reader or keyboard-only navigation.
+- Drag-and-drop builder: keyboard accessibility unknown. Drag-and-drop is typically inaccessible without explicit implementation.
+- The iOS Safari drag-and-drop bug (4 months open) likely also affects iOS AT users.
+- No accessibility statement on the website or in the app.
+
+---
+
+_No formal audit. WCAG compliance status unknown. The embedded form renderer ships to end-users' websites — any AT failures affect Canvas customers' users, not just Canvas users._

package/examples/product/demo/weeve/compass/compass-intake.md

@@ -0,0 +1,15 @@
+# compass-intake — Canvas
+# Source: sales notes, inbound inquiry log, retention data
+
+_Feed to `/compass-intake`._
+
+---
+
+- ICP today: SMB teams (5–50 people), non-technical users, form-heavy workflows.
+- Inbound is shifting: last 2 months, 8 demo requests from companies with 50–200 employees asking about SSO, audit logs, and API access. These aren't the current ICP.
+- One enterprise prospect (180 employees) walked away citing "no SAML." Estimated at $600/month.
+- Growth has slowed: MoM growth down from 12% to 6% over the last quarter. 1,200 active teams.
+- Retention pattern: high for teams using conditional logic. Low for teams using only basic forms — they often return to Google Forms.
+- Founding positioning: "Typeform for people who can't afford Typeform." Still the implicit frame — never formally updated.
+- Pricing: Free / $29 / $79. The $79 plan rarely comes up in customer conversations. Team doesn't know why.
+- No written ICP document. Founders describe the customer differently in different conversations.

package/examples/product/demo/weeve/felt/felt-intake.md

@@ -0,0 +1,37 @@
+# felt-intake — Canvas
+# Source: weeve/felt/user-research.md, NPS data, in-app analytics, CSAT survey
+
+_Feed to `/felt-intake`._
+
+---
+
+## Quantitative signal
+
+- NPS: 42 (March 2026, n=340)
+- Top promoter theme: "easy to set up, works for simple stuff"
+- Top detractor theme: "conditional logic is confusing" — in 5 of the last 6 qualitative NPS responses
+- CSAT on conditional logic: 2.8/5 (n=47, from in-app survey after using the feature)
+- 73% of users who try conditional logic abandon it in the first session
+- Most-visited page: response inbox (68% of all sessions include it)
+- Integrations page: only 12% of users have ever opened it
+
+## User interview themes (6 interviews, Feb–March 2026)
+
+- Conditional logic UX: 4 of 6 participants couldn't complete a conditional form in a task session. "I'd just make two separate forms."
+- Feature discovery: integrations and multi-page forms are invisible. Multiple users discovered them during the interview and reacted with surprise.
+- Delete with no confirmation: 1 participant visibly stressed while deleting a test form. "Are you sure there's no undo?"
+- Enterprise signal: 2 of 6 participants asked unprompted about SSO or audit logs. Neither is the target user type.
+- Mobile: 2 participants asked if the builder works on mobile. It doesn't — not communicated.
+
+## Behavioral data (analytics)
+
+- Avg forms per active team: 3.2
+- Teams who use conditional logic: 27% (low given it's a core feature)
+- Time to first form: median 11 minutes
+- Teams who import data via integrations: 12% (despite 6 integrations available)
+
+## What we don't know
+
+- Why teams churn — no churn interview process
+- What the 50–200 person company inbound actually needs vs what we assume
+- Whether conditional logic confusion is a UX problem or an onboarding problem (or both)

package/examples/product/demo/weeve/felt/user-research.md

@@ -0,0 +1,68 @@
+# User research notes — Canvas
+
+_Running doc. Most recent first._
+
+---
+
+## March 2026 — 6 moderated interviews
+
+**Recruiting:** existing free/pro users, mix of industries. 45-min sessions.
+**Tasks given:**
+1. Build a form with 3 conditional branches
+2. Find the Slack integration
+3. Delete a form (then try to recover it — there's no recovery)
+
+**Task completion:**
+
+| Task | Completed | Avg time | Notes |
+|---|---|---|---|
+| Conditional logic form | 2 of 6 | n/a | 4 gave up. "I'll just make two forms." |
+| Find Slack integration | 3 of 6 | 4.5 min | Not findable from main nav. |
+| Delete form | 6 of 6 | — | 1 visibly stressed. 2 asked "are you sure?" before clicking. |
+
+**Interview notes (selected):**
+
+*P1 — Marketing coordinator, 25-person agency:*
+"I gave up on the if/then stuff. I just make separate forms."
+Primary use: basic lead capture. Never returned to conditional logic after first try.
+
+*P2 — Operations manager, 80-person SaaS:*
+"It does what I need but I feel like I'm missing features I don't know about."
+Discovered Slack integration during the interview. Had been exporting to CSV for 6 months.
+Quote: "Why isn't this in the main nav?"
+
+*P3 — Freelance designer:*
+"The embedded form doesn't look great in my client's site. Styling options are limited."
+Workaround: CSS overrides in the client's CMS. Wants branded forms.
+
+*P4 — HR manager, 120-person company:*
+"I need to know who filled out what. The response inbox doesn't make that easy."
+Use case: employee onboarding. Wants respondent tracking and SSO.
+End of session: "Do you have a plan for enterprise?"
+
+*P5 — Product manager, 40-person startup:*
+"Conditional logic took me 3 tries. The 'then' panel is confusing."
+Specific issue: if/then panels are visually disconnected — no line or arrow showing the relationship.
+Also noted: "Delete form should confirm. I deleted the wrong one once."
+
+*P6 — Small business owner, e-commerce:*
+"I use it for customer feedback. It's fine. I just wish the emails looked better."
+Very low engagement. 1 form. Would switch if Google Forms added Slack.
+
+---
+
+## 2023 — Pre-launch research (3 interviews)
+
+Finding: users want "simple, not powerful." Key jobs: lead capture, events, feedback.
+Main concern: "will it look good on my website?"
+
+**Note:** These are 3 years old. Product has changed significantly. Personas from this research should not be used as current signal.
+
+---
+
+## Research backlog
+
+- [ ] Churned user interviews — why do low-engagement users leave?
+- [ ] Enterprise prospect interviews — 8 recent demo requests from 50–200 person companies
+- [ ] JTBD mapping — current understanding is demographic, not job-based
+- [ ] Accessibility testing with real AT users

package/examples/product/demo/weeve/layers-plus/layers-plus-intake.md

@@ -0,0 +1,14 @@
+# layers-plus-intake — Canvas
+# Source: team observations, user interview sessions, analytics
+
+_Feed to `/layers-plus-intake`._
+
+---
+
+- Conditional logic: "if" condition and "then" action are in separate panels with no visual connection. Users report not knowing which panel affects which.
+- Drag-and-drop builder communicates nothing about what's droppable — no hover states, no visual affordance. Users discover it by accident.
+- Response inbox: most-used feature (68% of sessions) but has had the fewest design iterations since MVP.
+- Empty states: none exist for new workspaces, empty form lists, or empty response inboxes.
+- Mobile: forms are responsive, builder is desktop-only with no messaging to users who try it on mobile.
+- Object model ambiguity: is a "form" a container for questions or a workflow? The product treats it as both — conditional logic implies workflow — but the UI only communicates "container."
+- Delete has no confirmation. Undo doesn't exist. Known gap; 2 data loss reports.

package/examples/product/demo/weeve/rubric/component-inventory.md

@@ -0,0 +1,42 @@
+# Component inventory — Canvas
+
+_Partial audit. Web app (builder + inbox) only._
+_Last updated: February 2026_
+
+## Components confirmed in use
+
+| Component | Location | Notes |
+|---|---|---|
+| Button (primary) | Throughout | Standalone file. |
+| Button (secondary) | Throughout | Copy of primary, different background. |
+| Button (danger) | Delete actions | Copy of primary, red. |
+| Input (light) | Builder, settings | |
+| Input (dark) | Embedded form renderer | Copy of light, inverted. |
+| Textarea | Builder | |
+| Select | Settings, logic builder | |
+| Checkbox | Builder | |
+| Radio | Builder | |
+| Toggle | Settings | |
+| Modal | Exists but... | Not used for delete confirmation — that's just a direct action. |
+| Card (plain) | Response inbox | |
+| Card (bordered) | Form list | Copy of plain with border. |
+| Badge | Form status labels | |
+| Spinner | Loading states | |
+| Toast | Success messages | Error toast: unclear if it exists |
+
+## Not inventoried
+
+- Conditional logic panels — built inline, not componentized
+- Field type selector — inline
+- Embedded form renderer — separate bundle, different styles, not reviewed
+
+## Missing (no component exists)
+
+- Empty states — needed in 4+ places
+- Confirmation dialog — needed for delete at minimum
+- Error state for form submission
+- Skeleton/loading for the response inbox
+
+---
+
+_6 confirmed copy-paste variants. No shared base components. No tokens._

package/examples/product/demo/weeve/rubric/rubric-intake.md

@@ -0,0 +1,14 @@
+# rubric-intake — Canvas
+# Source: weeve/rubric/component-inventory.md, code review
+
+_Feed to `/rubric-intake`._
+
+---
+
+- 22 components total. 6 are copy-pasted style variants with no shared base.
+- The 6 copy-paste components: Button (primary/secondary/danger), Input (light/dark), Card (plain/bordered). Each maintains its own padding and color values separately.
+- No design token system. Colors and spacing are hardcoded in component files.
+- No Storybook. Components are documented in a Notion page that's 8 months out of date.
+- No contribution process — engineers modify or add components without design review.
+- The conditional logic UI (most complex feature surface) is not componentized — built inline in a single 800-line file.
+- The embedded form renderer uses a separate stylesheet from the builder. They've diverged.

package/examples/product/demo/weeve/shared/competitive.md

@@ -0,0 +1,28 @@
+# Competitive notes — Canvas
+
+_Not a formal analysis. Based on what the team knows._
+
+## Direct competitors
+
+**Typeform** — premium UX, strong brand, expensive at scale. Mentioned in churn surveys.
+**Jotform** — feature-heavy, lower cost, cluttered. Popular with non-technical users.
+
+## Baseline
+
+**Google Forms** — the free comparison everyone makes. "It's like Google Forms but..."
+
+## What we hear from customers
+
+- "We switched from Typeform because it got too expensive" — main acquisition story
+- "Typeform looks better" — main design pressure
+- Nobody mentions Jotform in conversations; they've probably never encountered it
+
+## What we don't know
+
+- Why customers churn to Google Forms vs just canceling
+- What enterprise alternatives look like at the 100+ employee tier
+- Whether Notion Forms or similar newer tools are becoming relevant
+
+---
+
+_No formal win/loss tracking. Enterprise competitors (Formstack, Conga) not researched._

package/examples/product/demo/weeve/shared/roadmap.md

@@ -0,0 +1,32 @@
+# Roadmap — Canvas
+
+_Q3 2026 planning. Not all items are sized or committed._
+
+## Shipping this quarter (committed)
+
+- Multi-page forms — top feature request (3 explicit asks last month alone)
+- Conditional logic redesign — addressing the consistent UX complaint
+- iOS Safari drag-and-drop fix — 4 months overdue
+
+## Planned / in design
+
+- Delete confirmation dialog — 2 users have reported accidental data loss
+- Response inbox redesign — most-used feature, fewest design iterations
+- SSO (SAML) — enterprise inbound is asking; no timeline set
+
+## Backlog
+
+- API access (read responses programmatically)
+- Custom subdomain for embedded forms
+- Audit log — enterprise requirement
+- Native Salesforce/HubSpot integration (currently webhook-based)
+
+## Not on the roadmap
+
+- Mobile native app
+- Payments / Stripe integration
+- AI form generation
+
+---
+
+_Roadmap reviewed monthly. Backlog items shift. SSO has been "planned" for two quarters._

package/examples/product/demo/weeve/shared/weeve-guardrails.md

@@ -0,0 +1,27 @@
+# weeve-guardrails.md — Canvas
+
+## Hard limits
+- React frontend only — no migration
+- No native mobile apps — web responsive only
+- Integrations must use public APIs only — no screen scraping
+- GDPR compliance required (EU customers on paid plan)
+
+## Technical constraints
+- Drag-and-drop built on dnd-kit — switching libraries would require full rewrite
+- Embedded forms use iframe isolation — cannot share styles with host page
+- Webhooks are fire-and-forget (no retry mechanism yet)
+- Free plan is truly free — no credit card gate; limits enforced by quota
+
+## Resourcing
+- 1 PM, 2 designers, 5 engineers; no QA or research roles
+- No dedicated accessibility support
+
+## Non-negotiables
+- Public-facing forms must load in under 2 seconds (p75)
+- Response data must never be shared across tenant boundaries
+- The free plan must remain genuinely useful (no dark patterns to force upgrades)
+
+## Known UX debt
+- Conditional logic UI has been flagged as "confusing" in 5 of 6 user interviews
+- iOS Safari drag-and-drop bug has been open 4 months
+- "Delete form" is not guarded by a confirmation dialog — data loss has been reported twice

package/examples/product/demo/weeve/shared/weeve-intake.md

@@ -0,0 +1,18 @@
+# weeve-intake.md — Canvas observations
+
+---
+
+## Captured observations
+
+- NPS is 42; qualitative responses mention "conditional logic is confusing" in 5 of 6 recent replies
+- No formal user research since 3 interviews at launch 18 months ago
+- The drag-and-drop builder has a known regression on iOS Safari — unresolved for 4 months
+- "Delete form" has no confirmation dialog; two users have reported accidental data loss
+- 6 of 22 components are copy-pasted style variants with no shared base — will diverge further
+- Embedded form iframe was never tested with a screen reader or keyboard-only navigation
+- Conditional logic step adds cognitive load: users set "if" and "then" in separate panels with no visual connection
+- Inbound from 50–200 person companies requesting SSO and audit logs — not on the roadmap
+- One enterprise prospect walked away citing "no SAML" as a blocker
+- The response inbox is the most-used feature but has the fewest design iterations
+- Three users in the last month asked for "multi-page forms" — it's the top feature request
+- Color contrast on placeholder text fails WCAG AA in the default theme

package/examples/product/demo/weeve/shared/weeve-project.md

@@ -0,0 +1,29 @@
+# Canvas — project brief
+
+**What it is:** Canvas is a no-code form and workflow builder for SMBs. Teams build intake forms, approval flows, and lightweight internal tools without writing code. Think Typeform + Zapier but simpler, with better data ownership.
+
+**Stage:** Post-launch, 18 months in. ~1,200 active teams on free/pro plans. Starting to get inbound from companies with 50–200 employees wanting more customization and security controls.
+
+**Core product surface:**
+- Drag-and-drop form builder
+- Conditional logic engine (show/hide fields, branch workflows)
+- Response inbox and basic data table view
+- Integrations: Slack, Google Sheets, Airtable, webhooks
+- Public-facing forms + embedded forms (iframe)
+
+**Current state:**
+- The UX was designed by the founding engineer 18 months ago; no formal UX research has been done
+- 3 user interviews were run at launch; nothing systematic since
+- The component library has 22 components; 6 are copy-pasted variants with slight style differences
+- Embedded form accessibility is unknown — it was never tested with assistive technology
+- The drag-and-drop interaction has a known bug on iOS Safari (unresolved for 4 months)
+- User research: NPS is 42, but qualitative feedback consistently mentions "confusing conditional logic"
+
+**Team:**
+- 1 PM, 2 product designers, 5 engineers
+- No dedicated research, accessibility, or QA role
+
+**Open questions:**
+- Is the SMB ICP right, or is the inbound from 50–200 person companies telling us something?
+- The conditional logic UX has been called "confusing" in every user interview — redesign or patch?
+- Should we invest in a design system or keep shipping features?

package/examples/product/demo/weeve.yaml

@@ -0,0 +1,21 @@
+# weeve.yaml — product preset demo project
+schema_version: "0.4"
+
+project:
+  name: Canvas
+  tag: CVS
+  docs_root: ./weeve
+  depends_on:
+    - tag: VLT
+      name: Volta
+      relationship: platform  # Canvas runs on Volta's internal developer platform.
+                               # Volta's engineering health and API contracts affect Canvas's release readiness.
+
+roads:
+  - name: startup
+    path: roads/startup
+
+# Lanes in this preset: compass, layers-plus, ally, rubric, felt, pilot
+# Demo project: Canvas — no-code form and workflow builder
+# Depends on: Volta (engineering preset) — try running both to see cross-project signals in Pilot
+# Run /compass-intro to get started

package/examples/product/weeve.yaml

@@ -0,0 +1,14 @@
+# weeve.yaml — product preset
+# Generated by: weeve init product
+schema_version: "0.4"
+
+project:
+  name: My Project             # Replace with your project name
+  tag: MYP                     # Short ID — matches vault Projects/<tag>/ folder
+  # docs_root: ./weeve          # Uncomment to use a local path instead of vault
+
+roads:
+  - name: startup
+    path: roads/startup
+
+# Lanes in this preset: compass, layers-plus, ally, rubric, felt, pilot

package/examples/strategy/CODEOWNERS.template

@@ -0,0 +1,8 @@
+# Weeve lane ownership
+# Copy to .github/CODEOWNERS and replace team handles with your org's teams
+# Requires GitHub branch protection with "Require review from Code Owners" enabled
+
+weeve/compass/       @leadership
+weeve/layers-plus/   @product
+weeve/maven/         @marketing
+weeve/pilot/         @leadership