docs: record in the PRD what the evidence resolved

docs: platform and protocol findings reshape the plan

docs: delivery plan, engine/command dossiers, and the ADRs they ground

docs: PRD, ADR log and feasibility spikes for the multiplayer port

Merge pull request #95: fix the three engine defects the 3D client reported

Fix the three engine defects the 3D client reported (#92, #93, #94)

Merge branch 'claude/odyssey-background-planets-kzzw1t': bound the Odyssey living galaxy to 3 seed-drawn background worlds

Odyssey: bound the living galaxy to 3 seed-drawn background worlds

Fix the AI Editor tab: a blank screen from an undefined STRATEGIES

Merge branch 'claude/code-review-report-lyazay': repair the browser smoke job

Fix the smoke job: npx --package does not do what I assumed

Merge branch 'claude/code-review-report-lyazay': a browser smoke test in CI

Boot the game in a real browser in CI — the one thing 2502 tests cannot see

Merge branch 'claude/code-review-report-lyazay': close out the review follow-ups

Document the main-branch gate, and fix the docs my own changes falsified

Merge branch 'claude/code-review-report-lyazay': code review, and its fixes

Correct finding 3's impact claim — measured, it is narrower than I wrote

Cut 1.1.0, un-deprecate the CI actions, and run the browser smoke test

Split the two 800-line functions along the seams they already had

Close the guard gaps behind the last three bugs; withdraw one review finding

Get CI green, and make the AI's opponent belief fade the way it says it does

Add an external code review: CI is red, and the adaptive AI forgets too fast

Merge pull request #91: an in-app AI editor

An in-app AI editor: build an opponent dial by dial

Merge pull request #90: an adaptive AI, and a genome the player can own

The cast's labels do not survive held-out maps -- say so

The corrected cast: 10 cells from the median-descriptor archive

Record the adaptive-opponent work in the ledger and changelog

Phase 5: the roster can carry a player's own genome

Phase 6: play against yourself -- fingerprint a human into a genome

Phase 4: the adaptation POLICY is evolvable, not just its constants

Phases 2-3: the AI answers what it reads

Phase 1: the AI gets a belief about its opponent

Descriptors are a MEDIAN, and runSeed stops hashing the candidate's name

Calibrate the archive's health screen against the shipped AI, not a fixed list

Breed a CAST of distinct opponents, not one optimum: MAP-Elites

Merge pull request #89: evolve the AI — a genome, mutation, and crossover

Record the check guardrail result in the ledger row

Record what the first evolutionary run measured -- including that it fails

The first evolved candidate: 12 genomes x 6 generations, 3h21m

evolve: checkpoint the champion every generation, not only at the end

Evolve the AI: a genome, mutation, crossover, and a selection loop

Design brainstorm: how the AI could evolve by mutation and crossover

Merge pull request #88: in-game AI/human competitions with Elo ranking

Merge Phase 5: spectate, replay, seasons -- the plan completes

Phase 5 follow-ups: an undefined colour, an untrusted strategy, a stray "You"

Phase 5 fix: the ⌂ Home confirm was the third, un-closed door to Save

Phase 5 (replay + season): re-run any recorded match, and close a ladder

Phase 5 (spectate): watch an AI-vs-AI match live, at 1x-8x

Merge Phase 4: the human enters the field