No description
Find a file
jedarden 3b94b7eccb feat(matchmaker): add map fairness monitoring and auto-retirement (§14.6)
Implements the full map lifecycle audit as a hourly ticker in the
matchmaker:

1. updateMapFairnessStats: recompute per-slot win counts from completed
   matches into the map_fairness table
2. flagUnfairMaps: flag maps where any slot deviates >10pp from expected
   (1/N) across 80+ matches → status='probation'
3. retireDislikedMaps: force-retire maps with >20 net negative votes
4. pruneLowEngagementMaps: monthly bottom-10% engagement prune per tier
5. promoteClassicMaps: top-5 all-time engagement, 3+ months → 'classic'

Matchmaker already filters retired maps and gives probation maps 50%
reduced selection probability in selectMapLRU.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-04-22 18:19:24 -04:00
.github/workflows chore(ci): disable GitHub Actions — CI/CD is handled by Argo Workflows 2026-04-21 08:43:22 -04:00
bots feat(config): add season_id + rules_version to Config per §4.2 2026-04-22 18:09:26 -04:00
cmd feat(matchmaker): add map fairness monitoring and auto-retirement (§14.6) 2026-04-22 18:19:24 -04:00
docs feat(acb-api): implement bot registration, job coordination, and replay endpoints per plan §12 Phase 4 2026-04-21 08:58:42 -04:00
engine feat(config): add season_id + rules_version to Config per §4.2 2026-04-22 18:09:26 -04:00
manifests feat(evolver): add workflow completion polling to promoter 2026-04-22 17:46:33 -04:00
metrics feat(api): add token-bucket rate limiting to public endpoints 2026-04-22 16:52:29 -04:00
ratelimit feat(api): add token-bucket rate limiting to public endpoints 2026-04-22 16:52:29 -04:00
scripts feat(acb-api): implement bot registration, job coordination, and replay endpoints per plan §12 Phase 4 2026-04-21 08:58:42 -04:00
starters feat(config): add season_id + rules_version to Config per §4.2 2026-04-22 18:09:26 -04:00
web feat(config): add season_id + rules_version to Config per §4.2 2026-04-22 18:09:26 -04:00
.env.example docs: Update deployment guide for K8s + B2 architecture 2026-03-29 11:24:06 -04:00
.gitignore feat(playlists): add playlist curation and rebuild logic per §10, with series/seasons/enrichment 2026-04-21 16:11:27 -04:00
.needle-predispatch-sha feat(config): add season_id + rules_version to Config per §4.2 2026-04-22 18:09:26 -04:00
DEPLOYMENT.md feat(scripts): add R2 bucket setup script with custom domain config 2026-04-06 06:42:53 -04:00
DEPLOYMENT_STEPS.md docs: add deployment completion instructions 2026-04-08 16:55:20 -04:00
docker-compose.bots.yml feat(deploy): register 10 Phase 13 expansion bots in docker-compose.bots.yml 2026-04-22 17:55:56 -04:00
docker-compose.workers.yml Start Phase 6: Add deployment configuration and containers 2026-03-24 09:41:14 -04:00
go.mod feat(evolver): expand MAP-Elites from 2-D to 4-D grid per §10.2 2026-04-22 15:44:39 -04:00
go.sum feat(evolver): expand MAP-Elites from 2-D to 4-D grid per §10.2 2026-04-22 15:44:39 -04:00
PROGRESS.md docs: document series/season scheduler verification in PROGRESS.md 2026-04-21 16:39:43 -04:00
README.md Add README.md with project overview and quick start guide 2026-03-24 10:39:32 -04:00
wrangler.toml feat(ci): configure Cloudflare Pages build output directory 2026-04-08 17:07:37 -04:00

AI Code Battle

A competitive bot programming platform where participants write HTTP servers that control units on a grid world.

Overview

AI Code Battle is a game simulation platform where:

  • Participants write bots in any language that expose HTTP endpoints
  • Bots compete on a toroidal (wrapping) grid world
  • Matches are executed offline and presented as completed replays
  • A web platform shows leaderboards, match history, and replay viewers

Quick Start

Prerequisites

  • Go 1.21+ (for game engine and CLI tools)
  • Node.js 18+ (for web and worker components)
  • Docker (for containerized deployment)

Running Locally

# Build CLI tools
go build ./cmd/acb-local
go build ./cmd/acb-mapgen

# Run a match between built-in bots
./acb-local -seed 42 -max-turns 100 -output replay.json -verbose

# Start web development server
cd web && npm install && npm run dev
# Open http://localhost:3000/app.html

Viewing Replays

  1. Open the web app at http://localhost:3000/app.html
  2. Navigate to "Replay Viewer" in the menu
  3. Load a replay JSON file or enter a URL

Project Structure

ai-code-battle/
├── engine/              # Go game simulation library
│   ├── types.go         # Core data types
│   ├── grid.go          # Toroidal grid implementation
│   ├── game.go          # Game state management
│   ├── turn.go          # Turn execution phases
│   ├── replay.go        # Replay recording
│   └── *_test.go        # Test files
├── cmd/
│   ├── acb-local/       # CLI match runner
│   ├── acb-mapgen/      # Map generator
│   ├── acb-worker/      # Match execution worker
│   └── acb-indexer/     # Index builder for static files
├── web/                 # Cloudflare Pages SPA
│   ├── src/
│   │   ├── pages/       # Page components
│   │   ├── replay-viewer.ts  # Canvas replay renderer
│   │   └── app.ts       # SPA entry point
│   └── public/          # Static assets
├── worker-api/          # Cloudflare Worker API
│   └── src/
│       ├── index.ts     # Router + cron dispatcher
│       ├── jobs.ts      # Job coordination
│       ├── bots.ts      # Bot management
│       └── glicko2.ts   # Rating system
├── bots/                # Strategy bot implementations
│   ├── random/          # Python - RandomBot
│   ├── gatherer/        # Go - GathererBot
│   ├── rusher/          # Rust - RusherBot
│   ├── guardian/        # PHP - GuardianBot
│   ├── swarm/           # TypeScript - SwarmBot
│   └── hunter/          # Java - HunterBot
└── docs/plan/           # Implementation plan

Strategy Bots

Bot Language Strategy
RandomBot Python Random valid moves (baseline)
GathererBot Go Energy collection, avoid combat
RusherBot Rust Rush enemy cores aggressively
GuardianBot PHP Defend cores, cautious expansion
SwarmBot TypeScript Formation cohesion, group advance
HunterBot Java Target isolated enemies

Deployment

See DEPLOYMENT.md for detailed deployment instructions.

Quick Deploy

# Start all strategy bots
docker-compose -f docker-compose.bots.yml up -d

# Start match workers
docker-compose -f docker-compose.workers.yml up -d

Testing

# Go engine tests
go test ./engine/... -v

# Worker API tests
cd worker-api && npm test

# Index builder tests
cd cmd/acb-indexer && npm test

Architecture

The platform uses a split architecture:

  • Cloudflare (free tier): Static site, API endpoints, D1 database, R2 storage
  • Rackspace Spot: Match workers, bot containers, index builder

See docs/plan/plan.md for the full implementation plan.

License

MIT