20
prompts · 8 trigger pairs
+ install + skills + CLAUDE.md
Built for operators who'd rather hand-write a campaign than ship 'Don't Miss Our Spring Collection' at 7am with their BIMI logo next to it.
Klaviyo MCP install guide
Claude Desktop, Claude Code, ChatGPT Connectors, Gemini AI Studio - exact JSON config + API-key scoping rules.
Claude Skills configuration
The .claude/skills/klaviyo/ scaffold - brand voice doc, banned terms, approval policy referenced by every invocation.
CLAUDE.md governance template (~150 lines)
Voice rules, banned terms, approval gates, AI-SMS sample-match rules, jurisdiction-specific compliance (TX SB 140, VA SB 1339, TCPA).
20 production-grade prompts
Each runs against your live Klaviyo via MCP. Flow audit. Predictive readiness gate. Subject-line generator that loads your voice doc. AI-SMS pre-send classifier with cosine similarity. BIMI deployment walk-through. Yotpo post-migration cleanup. Full prompt text, copy-paste ready.
8 AI-flow trigger+prompt pairs
predicted-churn-soon, replenishment-window, browse-no-purchase, VIP-threshold-crossed, cart-abandoned-3hr, post-purchase-affinity, inactive-90d, high-CLV-no-purchase - for autonomous lifecycle.
Open / CTR reframe for the agent-to-agent inbox
Reply rate. CTOR. Inbox-placement rate. The composite engagement-signal score Apple/Gmail AI can't distort.
No brand voice document loaded as context
Composer’s Agent Guidance field accepts a tone document. Most operators leave it blank - and get cliche output. The 1,500-word voice doc template is in the pack.
No sample-match validation on AI-generated SMS
T-Mobile fines up to $10K per SHAFT-flagged content violation. AI improvises. Registered samples don’t. The pre-send classifier prompt is in the pack.
No predictive-readiness check before flow design
Klaviyo’s predictions need 500+ customers + 180+ days. Below the floor, every “predict and send” prompt produces garbage. The readiness gate is in the pack.
“The CLAUDE.md template is the actual artifact. Every agency stealing the rest of this content is missing that one file - and that's the one that prevents the 200-hallucinated-negatives failure.”
“The wasted-spend GAQL prompt found $4,200/mo in week one. Ran it on three more retainer accounts. $11.8K total recovered in 30 days.”
“I was paying $5–13 per Sora output for ad #20 of the week. The model-by-job matrix told me to use Kling for the bulk and Sora for hero. Cost dropped 78% the next month.”
“Showed the AI Slop Detector to our designer. We scored five live ads, three came back at 6+ flags. That's the 18% CPM tax we'd been paying without knowing.”
“Set up the Skills folder in 40 minutes. Every Claude session now loads with our Klaviyo context. Eliminated the 'paste in the voice doc' step every prompt used to need.”
“New hires get this on day one. The 12 GAQL prompts compress 6 months of 'how do you write that query again?' into one folder they can read on a flight.”