Changelog

Anthropic recently published a playbook called the AI-native SDLC. We have been adopting it heavily for HI Engine development.

The idea is simple. Code is no longer the bottleneck, so you rebuild the process around the parts that still run at human speed. Intent gets written down. Claude turns it into a spec, then a plan, then the change. Every PR is reviewed against rules we wrote once. Humans stop reading lines and start judging intent and risk.

It is working out nicely. We even have Arttu shipping code.

What we realized quickly is that HI Engine already runs this same loop, in principle, for CRO.

Intent comes from the data. Agents pull GA4, Shopify and ClickContext per store and emit findings that turn into CRO ideas. The merchant reviews the proposed change, the hypothesis, the evidence and a mockup — not a backlog ticket.

When the merchant likes the idea, the build runs from a queue into a draft theme, under rules that sit in the repo next to the rules for our own codebase. A verification gate screenshots every state on mobile and desktop before a human sees anything. If it fails, the draft is deleted and the merchant never knows it existed.

When the build is approved, the change is launched as an A/B test. Then the data decides. Revenue per visitor against a control. Winner baked into the theme, loser rolled back. The outcome is written down so the next discovery pass reads it. A store ends up with its own knowledge base: copy changes are dead here, imagery and price presentation move.

Same loop, one difference. A codebase has tests. A store has customers. So the merchant side has a stricter gate and a stricter definition of done. A shipped feature is judged by whether it works. A shipped storefront change is judged by whether it made money against a control.

In that one respect, the merchant side of the loop is the more mature one.

Chat directly with your store’s data. The dashboard agent answers questions about conversion metrics, cart abandonment, and opportunities in real time. Ask “Why did conversion drop last week?” or “What’s killing mobile AOV?” and get instant analysis grounded in live GA4 and Shopify metrics.

New

• Chat interface — ask natural questions and get AI-powered answers with supporting data

• Live GA4 & Shopify metrics — the agent queries real-time conversion rate, AOV, funnel drop-offs, sessions, and revenue

• Full insight & action access — the agent reads your complete CRO idea library and implementation tasks, sorted by severity

• Pinned goals (beta) — define store conversion targets and let the agent track progress toward them

• Instant context — personalized greeting on login before the agent finishes analysis

Fixes

• Action queries now properly return all items

Improvements

• Agent scoped to on-site conversion only — no traffic, SEO, or ads proposals

• Conversational voice refined for merchant-first framing and actionable recommendations

Why it matters

Insights are worthless without action. The agent turns raw data into a conversation partner who helps you move.

Every new CRO opportunity lands in Slack instantly, keeping your entire team synchronized without email overload.

New

• Per-store Slack channels — new ideas post automatically with a summary, estimated €/month impact, and a direct review link

• Smart deduplication — eliminates duplicate notifications when multiple related ideas generate within seconds

• Per-store kill switch — toggle Slack notifications on/off for individual stores without affecting email or other stores

• Clear attribution — all customer feedback and task notes properly marked as customer-authored vs. AI-generated

Improvements

• Simplified Slack headers for clarity

• Decoupled Slack from email logic — works regardless of store admin roster

• Removed screenshot spam; the preview link is more useful than static images

Why it matters

Your team moves fast. Slack keeps everyone in sync the moment a revenue opportunity is discovered.

Never miss a high-impact CRO opportunity. Instead of an email for every new idea, you get a digest that intelligently batches and prioritizes what matters most for your store.

New

• Twice-weekly digest — all new CRO ideas batch into a single email twice per week, replacing one-per-idea notifications

• Hard daily cap — never more than one email per store per day, even during high-insight periods

• Per-store controls — enable or disable email notifications for individual stores independently

• Smart prioritization — the most impactful ideas float to the top of each digest so you see the revenue opportunities first

Improvements

• Per-store kill switch lets you keep email on for some stores while switching to Slack-only for others

• Works alongside Slack — choose which channel fits each store best

Why it matters

Insights are only valuable if your team sees them. We optimized for signal over noise — thoughtful summaries instead of notification chaos.

More power and control over your implementation workflow — enhanced validation, workspace tools, and execution tracking.

New

• Shopify Theme Check — every modification validates against Shopify specs before publishing, catching schema violations and errors

• Per-task workspace — helper tooling for pulling themes, validating code, testing, and pushing changes

• Execution logs — full transparency into what changed, why, and how to roll back if needed

• Implementation status tracking — monitor task progress from approval through draft review to live publishing

Fixes

• Block ID collisions prevented when multiple patches apply simultaneously

• Lifecycle state guard prevents rejecting already-published tasks

Improvements

• Draft themes named `HI engine – <task>` for clarity and merchant context

• Theme Check catches real bugs, like min/max constraint violations, before they ship

Why it matters

Execution is complex. Better visibility and validation means fewer bugs and faster deployments.

HI Engine now automatically extracts brand profile knowledge from your live storefront on integration, giving the AI instant context about your business.

New

• Storefront auto-extraction — a two-pass agent investigates your store and structures findings into a complete brand profile

• Core profile data — automatically captures brand identity, tone, target customer, catalog shape, trust signals, and constraints

• Search-ready knowledge — all extracted knowledge is embedded and indexed for instant AI recall during analysis

• Idempotent rescans — re-run extraction anytime to refresh knowledge; manual edits and uploads are never overwritten

Fixes

• Non-core knowledge items now properly chunked and embedded for search

Improvements

• Security verification: knowledge access checks store membership before loading context

• Knowledge source tracking prevents automated data from overwriting manual inputs

Why it matters

The best recommendations come from understanding your business. Auto-extraction means the AI knows your store inside-out from day one.