Skip to content

New Agents: Inventory Forecasting + Landing Page Router (Wolf Tactical / Society Brands)#2

Open
SocietyDev wants to merge 1 commit intopaperclipai:mainfrom
SocietyDev:society-brands-agent-submissions
Open

New Agents: Inventory Forecasting + Landing Page Router (Wolf Tactical / Society Brands)#2
SocietyDev wants to merge 1 commit intopaperclipai:mainfrom
SocietyDev:society-brands-agent-submissions

Conversation

@SocietyDev
Copy link

Two E-commerce AI Agents from Project Autonomous Wolf

Company: Society Brands — 13-brand DTC portfolio (~$120M annual revenue)
Context: 54-day sprint (Feb 6 – Mar 31, 2026) to prove Wolf Tactical ($10M brand) can run with 2 humans + Paperclip AI agents


Agent #1: Inventory Forecasting Agent

Status: Phase 1 prototype (40% complete, production-ready framework)
Path: companies/society-brands-wolf-tactical/inventory-forecasting-agent/

Problem solved: Wolf Tactical is 75% Amazon — a stock-out means losing the Buy Box, which cascades to lost sales, ranking drops, and review suppression. Manual reorder tracking can't keep up with 172 SKUs.

What's included:

  • Production database schema (7 tables + 3 views) for multi-brand inventory tracking
  • Shopify Admin API sync script with rate limiting and error handling
  • Sales velocity calculator with rolling window averages
  • 37-page control plan + Phase 1 status report

Decision boundary: Agent forecasts and recommends; all reorder commitments escalate to human approval before execution.


Agent #2: Landing Page Router (Auto A/B Testing)

Status: 90% complete, LIVE at webapprouter.netlify.app
Path: companies/society-brands-wolf-tactical/landing-page-router/

Problem solved: Manual A/B test review cycles take 1-2 weeks. This kills losers automatically within 6 hours of hitting statistical significance.

What's included:

  • Netlify serverless router with weighted traffic splitting + segment targeting
  • GA4 Data API integration for conversion tracking
  • n8n auto-kill workflow (kills variants with <1.5% CVR after 200 sessions)
  • Process documentation + SOP PDF

What remains to fully activate: Connect ad platform traffic, run first Wolf Tactical test campaign.


Why These Matter for Paperclip Community

  1. Real-world production examples — not toy demos, actual business problems
  2. Clear governance model — autonomous monitoring + human approval before financial actions
  3. Multi-source integration patterns — Shopify, Amazon, GA4, n8n, Netlify, Paperclip
  4. Reusable for any e-commerce brand — parameterized for multi-brand support

Feedback Requested

  • Agent architecture review (are we thinking about decision boundaries correctly?)
  • Paperclip integration patterns (best way to wire approval flows, alerting?)
  • Community validation — useful for other e-commerce Paperclip users?

Contact: Dustin Brode — dustin.brode@societybrands.com

🤖 Submitted by Brand President agent (claude-sonnet-4-6) on behalf of Society Brands

Two real-world autonomous agents from Project Autonomous Wolf (54-day sprint
to prove $10M Wolf Tactical brand can run with 2 humans + Paperclip agents):

1. Inventory Forecasting Agent (40% complete)
   - Production DB schema (7 tables + 3 views)
   - Shopify sync + sales velocity calculator
   - Full control plan + Phase 1 status report
   - Escalation model: agent forecasts, human approves reorders

2. Landing Page Router / Auto A/B Testing (90% complete, LIVE)
   - Netlify serverless router with weighted traffic splitting
   - GA4 tracking + n8n auto-kill workflow (<1.5% CVR → kill after 200 sessions)
   - Process documentation + SOP PDF

Co-Authored-By: Claude (claude-sonnet-4-6) <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants