diff --git a/docs/consultations/gemini-pricing-tiers-round3-2026-04-13.md b/docs/consultations/gemini-pricing-tiers-round3-2026-04-13.md
new file mode 100644
index 0000000..056dbe7
--- /dev/null
+++ b/docs/consultations/gemini-pricing-tiers-round3-2026-04-13.md
@@ -0,0 +1,83 @@
+# Gemini Consultation: Pricing Tiers — Round 3
+
+**Date:** April 13, 2026  
+**From:** Michael (The Wizard) + Claude (Chronicler #85)  
+**To:** Gemini (Architectural Partner)  
+**Re:** Correction — ModpackChecker is NOT on the April 15 deadline
+
+---
+
+## Hey Gemini! 👋
+
+Really appreciate the Round 2 response — the Arbiter support model framing
+is exactly right and Meg will love hearing that confirmed. The egg injection
+risk assessment is also spot on.
+
+But we need to correct a fundamental assumption in our consultation.
+
+---
+
+## The Correction
+
+**ModpackChecker does NOT have to ship on April 15.**
+
+April 15 is the soft launch of **Firefrost Gaming** — the server community.
+That is a fixed date. ModpackChecker is a separate BuiltByBit product that
+ships when Michael is satisfied with it. That could be tonight, it could be
+next week, it could be later.
+
+Michael's exact words: *"I am not releasing this until I am satisfied,
+whether that is 6pm tonight or sometime in 2036."*
+
+This completely changes the risk calculus. We are not in a 48-hour trap.
+We have time to do this right.
+
+---
+
+## Revised Questions
+
+With the deadline pressure removed, please re-evaluate:
+
+1. **Egg Variable Auto-Injection** — Given that we now have time to build
+   and test it properly, does this become the centerpiece Pro feature that
+   we wait for before launching ModpackChecker on BuiltByBit? Or do we still
+   ship without it and add it as v1.2?
+
+2. **What does "done right" look like for the Pro tier?** If we have weeks
+   instead of 48 hours, what would you build before calling ModpackChecker
+   Professional ready to sell?
+
+3. **The rollback question you raised** — what IS the right rollback strategy
+   for a BuiltByBit release that uncovers a panel-breaking bug on a customer's
+   machine? This is worth answering regardless of timeline.
+
+4. **Sequencing** — Should we ship Standard first (stable, well-tested, clear
+   value) and launch Professional separately once egg injection is built? Or
+   launch both tiers simultaneously when Pro is complete?
+
+---
+
+## What We Have Right Now (stable)
+
+- Console widget — zero-click, ErrorBoundary, async error handling ✅
+- Truth File detection — never seeds from latest ✅
+- Calibration flow — pending_calibration + Identify Version button ✅
+- All platforms (CurseForge, Modrinth, Technic, FTB) ✅
+- Dashboard badges ✅
+- License system (standard/professional gating) ✅
+- Webhook + interval infrastructure (built, needs Pro gating) ✅
+
+## What's Not Done Yet
+
+- Egg Variable Auto-Injection (Pro flagship feature idea)
+- Dashboard badges gated to Pro only (needs implementation)
+- Card position change (above Address card)
+- Calibration dropdown rendering fix (pending Code)
+- Rate limit adjustment (pending Code)
+- BuiltByBit listings (need rewriting with honest copy)
+
+---
+
+Thanks Gemini! Take your time — we're not rushing anymore. 🔥❄️
+
+— Michael + Claude (Chronicler #85)