feat: add durable memory edge plan

.autoclaw/mateam/scratch/2026-06-10-homepage-strategy-spotless/context.md

@@ -0,0 +1,18 @@
+# Context
+
+Canonical file:
+- `docs/homepage_strategy.md`
+
+Existing supporting file:
+- `docs/revenue_homepage_strategy.md`
+
+Automation:
+- `mindreply-homepage-strategy-hourly-review`
+
+Subagent themes integrated:
+- Hero must sell one send-ready reply, not a broad concept.
+- GBP 600 pack must be scoped.
+- Growth triggers repeat overload.
+- Pro triggers team, sensitive volume, receipts, and review.
+- Trust needs a concrete receipt card.
+- Implementation spec needs CTA behavior, pricing cards, UI treatments, and event keys.

.autoclaw/mateam/scratch/2026-06-10-homepage-strategy-spotless/output.md

@@ -0,0 +1,17 @@
+# Output
+
+Implemented:
+- `docs/homepage_strategy.md`
+- updated `tests/executive_nervous_system_scaffold.py`
+- hourly heartbeat automation created: `mindreply-homepage-strategy-hourly-review`
+
+The canonical strategy includes:
+- revenue spine
+- copy rules
+- ruthless diagnosis
+- two-layer homepage strategy
+- hero, trust, how-it-works, pricing, Growth/Pro, authority
+- outbound DMs, cold emails, follow-ups, call booking, objections
+- first-session conversion logic
+- implementation checklist
+- CTA behaviors, card specs, receipt UI, proof strip UI, event keys, claim discipline

.autoclaw/mateam/scratch/2026-06-10-homepage-strategy-spotless/plan.md

@@ -0,0 +1,11 @@
+# MAteam Plan: Homepage Strategy Spotless
+
+Dispatch path: mixed. The host provided `multi_agent_v1` subagents, not a literal `Agent` tool. Twelve critique lanes were attempted in waves; final integration was performed in-session.
+
+Task: create a canonical, implementation-ready `docs/homepage_strategy.md`, preserve it with hourly review, and verify it with tests.
+
+Roles:
+- Researcher lanes: buyer-job clarity, Growth/Pro triggers, trust proof, outbound copy, conversion logic.
+- Reviewer lanes: structure, phrase guardrails, paid path, copy sharpness, implementation readiness, final QA.
+- Coder: in-session file creation and test updates.
+- Verifier: in-session test, scan, typecheck, build, commit, push.

.autoclaw/mateam/scratch/2026-06-10-homepage-strategy-spotless/review.md

@@ -0,0 +1,10 @@
+# Review
+
+Subagent blockers resolved:
+- Secondary CTA no longer pushes GBP 600 before proof.
+- Trust block uses protected-detail receipt proof rather than slogan-only language.
+- GBP 600 pack has scope.
+- Credits, pack, Growth, and Pro are separated by buyer moment.
+- Growth threshold is stricter than casual usage.
+- Pro triggers are tied to team review, sensitive risk, receipts, and repeated high-trust work.
+- CTA behavior, pricing cards, proof strip UI, trust card UI, how-it-works UI, and event keys are specified.

.autoclaw/mateam/scratch/2026-06-10-homepage-strategy-spotless/verify.md

@@ -0,0 +1,14 @@
+# Verify
+
+Completed on 2026-06-10.
+
+- Python scaffold tests: pass, `9 tests`.
+- Blocked public-term scan across `app`, `src`, `site`, `docs`, and `playbooks`: pass, no matches.
+- Public internal-phrase scan across `app` and `site`: pass, no matches.
+- TypeScript check: pass, `tsc --noEmit`.
+- Production build: pass, `next build`.
+
+Build summary:
+- Next.js `15.5.19`.
+- Homepage route `/`: static, `7.57 kB`, first load `110 kB`.
+- Dynamic routes retained for action, intake, locale, memory, and owner endpoints.

.autoclaw/mateam/scratch/2026-06-10-revenue-homepage-copy/context.md

@@ -0,0 +1,12 @@
+# Context
+
+Current branch: `codex/mindreply-moa-main`.
+
+Relevant files:
+- `docs/` holds strategy and policy artifacts.
+- `tests/executive_nervous_system_scaffold.py` scans public product sources for blocked terms and validates key product contracts.
+
+Subagent findings used:
+- Strongest live conversion hooks: "Reclaim 2+ hours daily within 24 hours", no long setup, first action queue or send-ready message, pack purchase after first output.
+- Main weakness: the page mixes internal category language with buyer-facing copy and does not make Growth/Pro pressure visible enough.
+- Required direction: sell immediate operational relief first, then reveal premium communication intelligence depth.

.autoclaw/mateam/scratch/2026-06-10-revenue-homepage-copy/output.md

@@ -0,0 +1,16 @@
+# Output
+
+Planned deliverables:
+- `docs/revenue_homepage_strategy.md`
+- `tests/executive_nervous_system_scaffold.py` coverage for all requested sections
+
+Implemented deliverables:
+- `docs/revenue_homepage_strategy.md`
+- `tests/executive_nervous_system_scaffold.py`
+
+The artifact includes:
+- Ruthless diagnosis
+- Two-layer homepage strategy
+- Hero, trust, how-it-works, pricing path, Growth/Pro upgrade, authority block
+- 5 outbound DMs, 3 cold emails, 2 follow-ups, 1 call booking message, objection handling
+- First-session conversion logic

.autoclaw/mateam/scratch/2026-06-10-revenue-homepage-copy/plan.md

@@ -0,0 +1,11 @@
+# MAteam Plan: Revenue Homepage Copy
+
+Dispatch path: mixed. The host exposed `multi_agent_v1` subagents, not a literal `Agent` tool. Several sidecar lanes ran through `multi_agent_v1`; final integration, review, and verification are handled in-session.
+
+Task: produce GitHub-backed MindReply revenue copy and conversion strategy for the live-site reality.
+
+Roles:
+- Researcher: collect live-positioning diagnosis and subagent findings.
+- Coder: create the revenue strategy artifact and coverage test.
+- Reviewer: check for missing requested sections and blocked public terms.
+- Verifier: run Python tests, typecheck, and build.

.autoclaw/mateam/scratch/2026-06-10-revenue-homepage-copy/review.md

@@ -0,0 +1,10 @@
+# Review
+
+No blockers found.
+
+Checks:
+- The artifact covers every requested deliverable.
+- Copy keeps immediate revenue pressure first and premium communication intelligence second.
+- Growth and Pro have explicit usage triggers.
+- Public-source blocked terms are avoided.
+- The artifact is a strategy/copy deliverable, not a live homepage implementation.

.autoclaw/mateam/scratch/2026-06-10-revenue-homepage-copy/verify.md

@@ -0,0 +1,7 @@
+# Verify
+
+Passed:
+- Product-source blocked-term scan: clean.
+- `python -m unittest tests.executive_nervous_system_scaffold`: 9 tests OK.
+- `tsc --noEmit`: clean.
+- `next build`: successful.

.dockerignore

@@ -0,0 +1,11 @@
+.next/
+node_modules/
+npm-debug.log
+.env
+.env.local
+.env.development.local
+.env.test.local
+.env.production.local
+.git/
+.gitignore
+README.md

.env.example

@@ -0,0 +1,20 @@
+NEXT_PUBLIC_SITE_URL=http://localhost:3000
+
+# Mail intake keeps sender, subject, and snippet by default.
+MAIL_SNIPPET_MODE=true
+GMAIL_CLIENT_ID=
+GMAIL_CLIENT_SECRET=
+IMAP_HOST=
+IMAP_PORT=993
+IMAP_USER=
+IMAP_PASSWORD=
+
+# Calendar actions are prepared until explicit connection is configured.
+CALENDAR_PROVIDER=
+CALENDAR_CLIENT_ID=
+CALENDAR_CLIENT_SECRET=
+CALENDAR_ICS_URL=
+
+# Local signing key used for development receipts only.
+MINDREPLY_DEVICE_KEY_ID=local-device
+MINDREPLY_DEVICE_SIGNING_KEY=

.gitattributes

@@ -0,0 +1,12 @@
+* text=auto
+
+*.sh text eol=lf
+*.yml text eol=lf
+*.yaml text eol=lf
+*.ts text eol=lf
+*.tsx text eol=lf
+*.js text eol=lf
+*.jsx text eol=lf
+*.json text eol=lf
+*.md text eol=lf
+Dockerfile text eol=lf

.github/workflows/decision-layer.yml

@@ -0,0 +1,42 @@
+name: Decision Layer Verification
+
+on:
+  push:
+    branches:
+      - main
+      - "codex/**"
+  pull_request:
+    branches:
+      - main
+
+jobs:
+  verify:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: "20"
+
+      - name: Install web dependencies
+        run: npm install
+
+      - name: Typecheck
+        run: npm run typecheck
+
+      - name: Build
+        run: npm run build
+
+      - name: Setup Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Compile Python
+        run: python -m compileall src tests
+
+      - name: Run backend tests
+        run: python -m unittest tests.executive_nervous_system_scaffold -v

.github/workflows/hourly-owner-report.yml

@@ -0,0 +1,55 @@
+name: Hourly Owner Report
+
+on:
+  schedule:
+    - cron: "0 * * * *"
+  workflow_dispatch:
+
+permissions:
+  contents: read
+
+jobs:
+  owner-report:
+    if: github.ref == 'refs/heads/main' || github.event_name == 'workflow_dispatch'
+    runs-on: ubuntu-latest
+    env:
+      MINDREPLY_REPORT_ENABLED: "true"
+      MINDREPLY_REPORT_DRY_RUN: "false"
+      MINDREPLY_REPORT_CHANNELS: "email,slack"
+      MINDREPLY_REPORT_EMAILS: "angellllkr@gmail.com,info@mind-reply.com"
+      MINDREPLY_REPORT_EMAIL: "angellllkr@gmail.com"
+      MINDREPLY_REPORT_FROM: ${{ secrets.MINDREPLY_REPORT_FROM }}
+      RESEND_API_KEY: ${{ secrets.RESEND_API_KEY }}
+      MINDREPLY_SLACK_WEBHOOK_URL: ${{ secrets.MINDREPLY_SLACK_WEBHOOK_URL }}
+      SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
+      VERCEL_DEPLOY_STATUS: ${{ vars.VERCEL_DEPLOY_STATUS }}
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: "20"
+
+      - name: Install dependencies
+        run: npm install
+
+      - name: Check report lane
+        run: npm run report:check
+
+      - name: Build owner report
+        run: npm run launch:report
+
+      - name: Audit owner blueprint
+        run: npm run audit:blueprint
+
+      - name: Send owner report
+        run: npm run report:send
+
+      - name: Upload owner evidence
+        if: always()
+        uses: actions/upload-artifact@v4
+        with:
+          name: mindreply-owner-report
+          path: .mindreply/reports/**

.gitignore

@@ -1,30 +1,33 @@
-logs
-*.log
-npm-debug.log*
-yarn-debug.log*
-yarn-error.log*
-pnpm-debug.log*
-lerna-debug.log*
+# dependencies
+node_modules/
+/.pnp
+.pnp.*
 
-node_modules
-dist
-dist-ssr
-*.local
+# testing
+/coverage
 
-.vscode/*
-!.vscode/launch.json
-!.vscode/extensions.json
-.idea
-.DS_Store
-*.suo
-*.ntvs*
-*.njsproj
-*.sln
-*.sw?
+# next.js
+/.next/
+/out/
 
-/.cache
+# production
 /build
-.env*
-*.vars
-.wrangler
-_worker.bundle
+
+# env
+.env
+.env.local
+.env.*.local
+!.env.example
+
+# debug
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+dev-server*.log
+preview-server*.log
+
+# typescript
+*.tsbuildinfo
+next-env.d.ts
+.vercel
+.mindreply/