Release History

This file is generated by bun run versions:update. Do not edit manually—update CHANGELOG.md instead.

• 0.197.11 – 2025-11-30 (changelog)

  • Type Guards Module: Added packages/bot/src/runtime/types/typeGuards.ts with type guard utilities (#1565)
  • isCreep(), isSource(), isStructure(), isSpawn(), isContainer(), isTower(), etc. type guards for runtime validation
  • asCreep() validation helper that throws TypeError instead of silently passing invalid objects
  • Type-safe room.find() wrappers: findActiveSources(), findAllSources(), findMySpawns(), findHostileCreeps(), findMyCreeps(), findDroppedResources(), findContainers(), findTowers(), etc.
  • Added 24 unit tests for type guards functionality
  • Behavior Controllers: Reduced unsafe type assertions from 50+ to ~10 instances (#1565)
  • Replaced creep as Creep assertions with validated asCreep() helper in 14 role controllers
  • Replaced scattered as Source[], as StructureSpawn[], etc. with centralized type-safe helpers
  • Updated helpers.ts to use type guard utilities instead of inline assertions
  • Improved code maintainability by centralizing type assertions in one location

• 0.189.9 – 2025-11-29 (changelog)

  • PTR Stats Collection Enhancement: Restored and enhanced PTR stats collection for performance validation and strategic analysis (#1549)
  • Added historical trend analysis script (analyze-historical-trends.ts) for tracking 7-day and 30-day performance trends
  • Implemented regression detection with configurable thresholds for CPU usage, bucket health, and resource metrics
  • Created comprehensive PTR stats schema documentation (packages/docs/source/docs/analytics/ptr-stats-schema.md) for strategic planning agents
  • Enhanced profiler data collection with improved error handling for “undefined” console responses
  • Added reports/monitoring/historical-trends.json to track performance trends across multiple time periods
  • Integrated trend analysis into monitoring workflow with automatic artifact generation
  • Added 20 unit tests for historical trend analysis functionality
  • Enables data-driven optimization and autonomous performance monitoring
  • Unblocks strategic planning agent performance analysis capabilities
  • Supports multi-shard telemetry collection (MMO + PTR)
  • Provides baseline metrics for optimization validation and risk mitigation

• 0.189.7 – 2025-11-29 (changelog)

  • Monitoring: Fixed bot snapshot collection to use multi-shard console telemetry (#1503)
  • Enhanced collect-bot-snapshot.ts to collect data from all shards where bot has rooms
  • Added multi-shard console API queries instead of relying on single-shard Stats API
  • Implemented snapshot validation to detect stale/empty data before committing
  • Added comprehensive diagnostic logging for troubleshooting collection failures
  • Snapshots now include per-room shard metadata for accurate multi-shard tracking
  • Validation fails workflow if critical data quality issues detected (0 creeps with claimed rooms)
  • Prevents false positive alerts from incomplete single-shard data collection

• 0.189.3 – 2025-11-29 (changelog)

  • Task System: Fixed cross-room over-spawning issue where tasks from multiple rooms caused excessive creep spawning
  • Added roomName field to TaskQueueEntry to track which room each task belongs to
  • Updated task discovery functions to include room name in task IDs (e.g., W1N1-build-site-123)
  • Added getTaskCountForRoom() method to RoleTaskQueueManager for room-specific task counting
  • Modified spawn scaling logic to only count tasks from the spawn’s room when calculating workforce needs
  • This prevents scenarios where builders are spawned in Room A due to construction sites in Room B

• 0.187.9 – 2025-11-29 (changelog)

  • Dismantler: Fixed dismantler attempting to dismantle novice zone walls near room edges
  • Added isNearRoomEdge() helper to detect structures within 1 tile of room boundaries (x ≤ 1, x ≥ 48, y ≤ 1, y ≥ 48)
  • Structures at room edges are now filtered out in both combat and clearing modes
  • This prevents wasted time targeting indestructible novice/respawn zone walls that block exits

• 0.183.5 – 2025-11-28 (changelog)

  • Documentation: Consolidated dual documentation locations into single Hexo site
  • Migrated all content from packages/docs/docs/ to packages/docs/source/docs/
  • Removed legacy packages/docs/docs/ directory
  • Updated workflow references from docs/ to packages/docs/source/docs/
  • Fixed guard-version.yml to check correct changelog path
  • Fixed guard-deprecation.yml to link to correct deprecation policy location
  • Fixed deploy.yml to reference correct troubleshooting documentation
  • Fixed performance-test.yml to reference Docker development guide
  • Updated AGENTS.md to remove obsolete “NOT root docs/“ warnings
  • Migrated 43 missing documentation files including:
  • Development guides: deprecation-policy.md, deprecation-registry.md, changelog-format.md
  • Operations docs: cpu-timeout-incidents.md, performance-optimization.md, monitoring-alerts-playbook.md
  • Runtime docs: bootstrap.md, kernel-protocols.md, task-system.md, creep-communication.md
  • Strategy docs: architecture.md, phase implementation guides
  • External analysis: jon-winsley-analysis.md, screeps-quorum-analysis.md
  • Resolves issue ralphschuler/.screeps-gpt#1526 (consolidate dual documentation locations)

• 0.180.0 – 2025-11-28 (changelog)

  • Deployment: Version history tracking for reliable rollback
  • Added reports/deployments/deployment-history.json to track validated deployments
  • Created manage-deployment-history.ts script for recording and querying validated versions
  • Created types/deployment-history.ts with type definitions for deployment history
  • Workflow now records validated deployments after successful health checks
  • Rollback now uses deployment history to find last validated version (not just git tags)
  • History maintains last 5 validated deployments with validation metrics
  • Falls back to git tags if history is empty (backwards compatible)
  • Prevents infinite rollback loops by only recording successful deployments
  • Added unit tests and regression tests for deployment history tracking
  • Updated deployment rollback documentation with new rollback process
  • Resolves issue ralphschuler/.screeps-gpt#1496 (rollback lacks version history tracking)

• 0.176.0 – 2025-11-28 (changelog)

  • Automation: Custom Copilot agent environment configuration
  • Created .github/copilot-environment.json for project-specific agent setup
  • Configured Node.js 22 Alpine base image with Yarn 4.11.0
  • Set up @ralphschuler GitHub Package Registry authentication
  • Defined available tools (node, npm, yarn, git, gh, tsx, vitest, eslint, prettier)
  • Enabled TypeScript 5.4.3 with strict mode, ESLint, Prettier, and Vitest
  • Configured workspace structure for monorepo support
  • Added comprehensive documentation in packages/docs/source/docs/automation/copilot-environment.md
  • Enhanced .devcontainer/devcontainer.json for consistency with Copilot environment
  • Resolves issue ralphschuler/.screeps-gpt#1354 (enhance Copilot integration with custom environment)

• 0.175.7 – 2025-11-28 (changelog)

  • Pathfinding in Narrow Passages: Creeps now properly navigate through single-tile corridors by ignoring other creeps in pathfinding calculations
  • Added ignoreCreeps: true to all pathfinding operations (findClosestByPath, moveTo) across all role controllers
  • Updated helper functions in helpers.ts: findClosestOrFirst, tryPickupDroppedEnergy, moveToTargetRoom
  • Updated role controllers: HarvesterController, HaulerController, UpgraderController, BuilderController, RepairerController
  • Paths are now calculated based on terrain and structures only, not temporary creep positions
  • Prevents pathfinding failures when passages are temporarily occupied by other creeps
  • Creeps will still physically collide at runtime; traffic management handles actual collision avoidance
  • Added comprehensive unit tests for findClosestOrFirst helper validating ignoreCreeps behavior
  • Updated regression tests in room-exit-crossing.test.ts to expect ignoreCreeps: true
  • Resolves issue ralphschuler/.screeps-gpt#1466 (creeps stuck in narrow passages)
  • Addresses issue ralphschuler/.screeps-gpt#1450 (pathfinding not working in single-tile corridors)

• 0.174.0 – 2025-11-28 (changelog)

  • Automation: Enhanced Copilot prompts with Screeps MCP server documentation
  • Updated AGENTS.md to document Screeps Docs MCP and Screeps Wiki MCP servers
  • Added MCP tool references to all relevant prompts: researcher, issue-triage, strategist, strategic-planner, repository-audit, repository-review, todo-issue, todo-automation, screeps-monitor
  • Prompts now instruct agents to use screeps_docs_search, screeps_docs_get_api, screeps_docs_get_mechanics, screeps_wiki_search, and screeps_wiki_get_article for game knowledge
  • Enables automated workflows to gather Screeps game context when handling issues and implementing features

• 0.165.1 – 2025-11-26 (changelog)

  • Automation: Replaced copilot-exec custom implementation with wrapper around codex-exec (OpenAI’s official codex-action)
  • Reduced maintenance burden by leveraging official OpenAI tooling
  • Maintained full backward compatibility with all existing workflows (4 workflows, 5 specialized agents)
  • copilot-exec now delegates to codex-exec which wraps openai/codex-action@v1
  • Updated regression tests to validate delegation and compatibility
  • Updated documentation in packages/docs/source/docs/automation/overview.md, AGENTS.md, and .github/copilot-instructions.md
  • Migration aligns with zero-tolerance policy for obsolete code while preserving existing workflow integrations

• 0.161.3 – 2025-11-25 (changelog)

  • CI/CD: Consolidated guard workflows using strategy matrices for faster execution
  • Merged guard-lint.yml, guard-format.yml, and guard-yaml-lint.yml into guard-code-quality.yml with matrix strategy (3 parallel jobs)
  • Merged guard-test-unit.yml, guard-test-e2e.yml, guard-test-regression.yml, and guard-test-docs.yml into guard-tests.yml with matrix strategy (4 parallel jobs)
  • Updated quality-gate-summary.yml to monitor consolidated workflows
  • Reduced from 12 to 7 guard workflows (42% reduction)
  • Benefits: Parallel execution within matrix, faster CI feedback, reduced workflow file overhead
  • Updated documentation in packages/docs/source/docs/automation/overview.md and legacy docs

• 0.159.1 – 2025-11-25 (changelog)

  • Documentation: TSDoc/JSDoc documentation standards and style guide
  • Created packages/README.md with comprehensive TSDoc standards and templates
  • Added code documentation standards section to AGENTS.md and DOCS.md
  • Added ESLint JSDoc validation rules for exported classes (set to warn for incremental adoption)
  • Enhanced TSDoc comments for high-priority entry points (main.ts, server.ts, agent.ts)
  • Installed eslint-plugin-jsdoc for automated documentation validation

• 0.157.5 – 2025-11-25 (changelog)

  • Documentation: Comprehensive Screeps wiki topic research
  • Created docs/research/screepspl-wiki-comprehensive-topics.md covering all major wiki.screepspl.us topics
  • Documented 17 major topic areas: RCL progression, creep body parts, combat strategies, CPU optimization, minerals/labs, pathfinding, market economy, private servers, power mechanics, remote harvesting, links, observers, spawning, safe mode, room claiming, factories, and API constants
  • Added implementation priority matrix for .screeps-gpt integration
  • Cross-referenced current bot implementation status with wiki best practices
  • Supplements existing screepspl-wiki-analysis.md with detailed topic coverage
  • Documentation: Created GitHub Models in Actions migration assessment
  • Comprehensive comparison between Copilot CLI and actions/ai-inference
  • Identified workflows suitable for migration vs. those requiring Copilot CLI
  • Proposed hybrid architecture combining both approaches
  • Documented implementation plan with phased rollout
  • Added example showing issue classification pre-filter pattern
  • Located at packages/docs/source/docs/automation/github-models-migration.md

• 0.155.21 – 2025-11-25 (changelog)

  • Pathfinding: Integrated screeps-pathfinding library by NesCafe62
  • Added NesCafePathfinder as the default pathfinding provider
  • Added traffic management with priority-based movement (runMoves())
  • Added moveOffRoad() to move creeps off roads when finished working
  • Added reservePos() for position reservations
  • Added priority option in moveTo() for traffic management
  • Added move off exit behavior
  • Added terrain and cost matrix caching
  • Pathfinding: Replaced DefaultPathfinder and CartographerPathfinder with NesCafePathfinder
  • Automatic fallback to native Screeps pathfinding when library is unavailable
  • Pathfinding: Removed DefaultPathfinder implementation
  • Pathfinding: Removed CartographerPathfinder implementation

• 0.154.0 – 2025-11-24 (changelog)

  • Memory Management: Implemented proper memory usage patterns from wiki.screepspl.us/Memory
  • Added GlobalCache for volatile heap caching (no serialization overhead)
  • Added path serialization utilities (serializePath, deserializePath, etc.) for 85% memory reduction
  • Added flag memory cleanup to MemoryGarbageCollector
  • Updated memory management documentation with advanced patterns
  • Covers heap caching, path serialization, and RawMemory segments documentation

• 0.151.1 – 2025-11-24 (changelog)

  • Claimer Room Cycling: Fixed claimers cycling back through room exits after entering target room
  • When creep enters target room at edge position (x/y = 0 or 49), it now moves toward room center before attempting to claim
  • Prevents pathfinding to controller from leading creep back through the exit
  • Added regression tests to verify behavior at all edge positions and corners

• 0.146.0 – 2025-11-24 (changelog)

  • Build Validation: Added comprehensive test coverage for context-aware size thresholds
  • Validates main.js requires minimum 50KB (ensures functional AI with kernel + runtime components)
  • Validates module files require minimum 500 bytes (allows small type-only modules)
  • Tests confirm validation prevents broken or incomplete builds from passing
  • Added 11 regression test cases covering threshold enforcement and edge cases
  • Addresses deployment safety concerns from ralphschuler/.screeps-gpt#729

• 0.143.5 – 2025-11-24 (changelog)

  • Video Pipeline: Implemented Screeps room history video rendering infrastructure (stub implementation)
  • Created video pipeline configuration system with type definitions
  • Added capture-room-history.ts script for fetching Screeps replay data via API
  • Added render-room-video.ts script for video generation (stub, requires ffmpeg integration)
  • Added upload-to-youtube.ts script for YouTube upload automation (stub, requires googleapis)
  • Created GitHub Actions workflow .github/workflows/screeps-video-render.yml for automated pipeline
  • Added configuration example file config/video-pipeline.example.json
  • Added comprehensive documentation at docs/automation/video-rendering-pipeline.md
  • Pipeline supports configurable time windows, video quality settings, and YouTube metadata
  • Infrastructure complete, core rendering features require additional dependencies
  • Resolves issue ralphschuler/.screeps-gpt#1277

• 0.139.3 – 2025-11-24 (changelog)

  • Tower Repair: Fixed towers not repairing structures when no hostile creeps present
  • DefenseCoordinator now calls TowerManager.run() unconditionally
  • Towers will repair damaged infrastructure (roads, containers, etc.) during normal operations
  • Energy efficiency maintained (towers still reserve 500 energy for defense)
  • Added test coverage for repair-only scenarios
  • Resolves issue ralphschuler/.screeps-gpt#1275

• 0.137.21 – 2025-11-23 (changelog)

  • Workflow: Fixed screeps-agent.yml using npm instead of yarn, breaking workspace dependency resolution
  • Replaced npm ci + npm install with yarn install --frozen-lockfile
  • Updated Node.js version to use .nvmrc instead of hardcoded version
  • Aligns with other workflows (guard-build.yml, deploy.yml)
  • Resolves issue ralphschuler/.screeps-gpt#1270

• 0.137.15 – 2025-11-23 (changelog)

  • Refactoring: Extracted hardcoded spawn threshold constants from BehaviorController
  • Created packages/bot/src/runtime/behavior/constants.ts with SPAWN_THRESHOLDS export
  • Extracted CPU safety margin (0.85), energy reserve ratios (0.2, 50), storage thresholds (0.5, 0.3), and energy thresholds (0.9, 0.75, 0.8)
  • Updated BehaviorController.ts to import and use constants instead of magic numbers
  • Added comprehensive TSDoc documentation explaining threshold rationale
  • Added regression test validating constant values
  • No behavioral changes - all values remain functionally identical
  • Unblocks issue #1267 (state machine migration) by providing reusable threshold constants

• 0.137.9 – 2025-11-23 (changelog)

  • Breaking: Removed deprecated GitHub labels (bug, enhancement, severity/*)
  • Use type/bug, type/enhancement, and priority/* labels instead
  • No backward compatibility maintained as this is a single-user repository
  • Breaking: Removed Logger backward compatibility wrapper (packages/bot/src/runtime/utils/Logger.ts)
  • Import directly from @ralphschuler/screeps-logger package instead
  • Breaking: Removed deprecated getTaskQueue() function from BehaviorController
  • Function was unused and marked for removal
  • Updated documentation to remove backward compatibility references
  • Removed legacy label migration sections from label-system.md
  • Updated AGENTS.md to remove references to deprecated labels
  • Updated .github/copilot-instructions.md to remove backward compatibility notes

• 0.137.3 – 2025-11-23 (changelog)

  • Repository Cleanup: Removed unused files and placeholder packages to improve maintainability
  • Removed stale source/docs directory (migrated to packages/docs/source/ previously)
  • Removed placeholder package packages/actions (no implementation, marked as “future” in README)
  • Removed placeholder package packages/console (no implementation, marked as “future” in README)
  • No breaking changes - all builds, tests, and deployments remain functional
  • Workspace configuration automatically adapts to removed packages

• 0.125.7 – 2025-11-21 (changelog)

  • Workflow Architecture Refactoring: Simplified monitoring and CI automation workflows
  • Monitoring Workflow: Removed copilot-exec strategic analysis step from screeps-monitoring.yml
  • Workflow now focuses purely on data collection (PTR stats, bot snapshots, telemetry, profiler data)
  • Eliminates unnecessary AI analysis complexity from data pipeline
  • All data collection steps remain intact and functional
  • CI Automation: Replaced copilot-ci-autofix.yml with ci-auto-issue.yml
  • New workflow creates GitHub issues for failed CI runs instead of attempting automatic fixes
  • Preserves circuit breaker pattern (3 consecutive failures, 15-minute backoff)
  • Issues include workflow details, failed jobs, and links to run logs
  • Applies labels: automation, ci-failure, type/bug, priority/high, state/pending
  • Escalation issues created when circuit breaker trips
  • More predictable and transparent than automatic fixing
  • Removed copilot-ci-autofix-agent action and ci-autofix prompt
  • Updated test suites to reflect new workflow structure
  • Updated documentation (AGENTS.md, README.md) to describe new approach
  • Separation of concerns: Data collection vs. analysis
  • Predictability: Issue creation is more predictable than auto-fixing
  • Transparency: Issues provide visible tracking of CI problems
  • Maintainability: Simpler workflows are easier to debug

• 0.125.1 – 2025-11-20 (changelog)

  • Package Manager Migration: Migrated from Bun to Yarn 4 (Berry) as the primary package manager
  • Replaced bun.lock with yarn.lock for dependency management
  • Updated all documentation (README.md, DOCS.md, AGENTS.md, .github/copilot-instructions.md) to use yarn commands
  • Removed Bun-specific dependencies (bun-plugin-screeps)
  • Configured Yarn with node-modules linker for better compatibility
  • Added GitHub Package Registry configuration for @ralphschuler scope in .yarnrc.yml and .npmrc
  • Updated repository to be private for workspace support
  • All builds, tests, and workflows remain fully functional
  • Spawn Utilization Optimization: Lowered RCL 1-2 energy threshold from 90% to 80% for dynamic upgrader scaling
  • Fixes spawn idle issue where spawns remained inactive despite 82% energy capacity and significant CPU headroom
  • Enables proactive spawning of additional upgraders during healthy energy states (80%+ capacity)
  • Expected improvements: 20-30% faster RCL progression at early levels, increased spawn uptime from ~60% to >80%
  • Only affects RCL 1-2 (by design), higher RCLs use different thresholds
  • Added regression test suite validating spawn behavior at various energy thresholds (75%, 80%, 82%)
  • Resolves issue #1105: PTR spawn idle during healthy energy state
  • Screeps-Kernel Integration: Refactored bot package to use @ralphschuler/screeps-kernel decorator-based process management system
  • Replaced custom createKernel() factory with Kernel class from @ralphschuler/screeps-kernel
  • Created MainProcess wrapper using @process decorator to bridge screeps-kernel with existing bot logic
  • Renamed process decorator import to registerProcess to avoid shadowing Node.js global
  • Migrated from process.env to build-time constants (__PROFILER_ENABLED__, __TASK_SYSTEM_ENABLED__, __ROOM_VISUALS_ENABLED__)
  • Updated esbuild configuration to inject environment variables as compile-time constants
  • Enhanced screeps-profiler with null safety checks for Memory.profiler
  • Added comprehensive architecture documentation in SCREEPS_KERNEL_INTEGRATION.md
  • Zero breaking changes - all 783 unit tests pass
  • Establishes foundation for future modularization of bot components
  • Package Build Automation: Added postinstall scripts to all workspace packages
  • Automatically builds packages after npm install or bun install
  • Ensures packages are ready for import in workspace without manual build steps
  • Affects: screeps-kernel, screeps-profiler, screeps-logger, screeps-cache, screeps-perf, screeps-metrics, screeps-async, screeps-xstate, screeps-xtree, screeps-mcp, screeps-agent
  • Screeps Metrics Package: New package @ralphschuler/screeps-metrics for comprehensive metrics collection
  • Collects CPU metrics (usage, limits, bucket, tick limits, shard limits)
  • Monitors V8 heap statistics via Game.cpu.getHeapStatistics()
  • Tracks GCL (Global Control Level) progress
  • Tracks GPL (Global Power Level) progress when available
  • Collects detailed room-level metrics (energy, creeps, structures, hostiles)
  • Tracks resource availability (credits, pixels, CPU unlocks, access keys)
  • Fully typed with TypeScript
  • Configurable metrics collection for CPU optimization
  • Zero external dependencies (uses only official Screeps APIs)
  • Comprehensive test coverage with unit tests
  • Located at packages/screeps-metrics/
  • PTR Energy Capacity Monitoring Validation: Documented resolution of issue ralphschuler/.screeps-gpt#1056
  • Issue: Energy capacity utilization at 29.5% (133/450) during active extension construction
  • Root cause: Temporary resource constraint while 2 extension sites under construction
  • Resolution: Self-resolved through normal game progression when extensions completed
  • Outcome: Energy capacity increased from 450 to 500, achieving 100% utilization
  • Added regression test suite validating monitoring system correctly tracks energy trends during construction phases
  • Confirms low energy during construction is expected behavior, not a bug
  • Validates monitoring system’s ability to detect and track issue resolution automatically

• 0.116.0 – 2025-11-19 (changelog)

  • Automated Threat Detection and Defense Coordination System: Implemented comprehensive defense automation with threat detection, defensive posture management, and coordinated responses
  • Created ThreatDetector class for room scanning and threat assessment with intelligent scoring
  • Created DefenseCoordinator class to manage defensive responses and coordinate tower/combat operations
  • Integrated defense system into kernel execution flow (runs before infrastructure management)
  • Added defensive posture system: normal → alert → defensive → emergency
  • Threat levels: none, low, medium, high, critical (based on hostile count and body composition)
  • Threat scoring aligned with CombatManager: attackers (10 pts), healers (8 pts), ranged (5 pts), dismantlers (5 pts)
  • Memory persistence in Memory.threats, Memory.defense, Memory.combat
  • Automatic spawn priority adjustment: defenders prioritized during defensive/emergency postures
  • Controller upgrading pauses during defensive/emergency postures (issue ralphschuler/.screeps-gpt#702)
  • Upgraders move to safe positions near storage/spawns during combat
  • Memory cleanup for stale threat data (default: 100 ticks)
  • Added 23 comprehensive unit tests covering threat detection and defense coordination
  • Added comprehensive documentation in docs/runtime/defense.md
  • Resolves issue ralphschuler/.screeps-gpt#1024 (implement automated threat detection and defense coordination)
  • Multi-room Construction Budget: Enhanced ConstructionManager to support per-room construction limits
  • Added maxSitesPerRoom parameter (default: 1) to prevent single-room spam
  • Increased maxSitesPerTick default from 1 to 5 for multi-room scaling
  • Allows up to 5 rooms to build concurrently instead of 1 per tick
  • Scales better for multi-room operation (5x improvement)
  • Added regression test suite for multi-room construction planning
  • Resolves issue ralphschuler/.screeps-gpt#632 (ConstructionManager maxSitesPerTick=1 bottlenecks multi-room expansion)
  • Path Caching System: Implemented comprehensive path caching with TTL and LRU eviction for CPU optimization
  • Created PathCache class with configurable TTL (default 1500 ticks) and LRU eviction
  • Integrated path caching into DefaultPathfinder and CartographerPathfinder
  • Added cost matrix caching for terrain and structures
  • Cache metrics tracking: hit rate, CPU savings, cache sizes, evictions
  • Cache invalidation API for room and structure changes
  • Expected CPU savings: 20-30% reduction in pathfinding overhead
  • Expected pathfinding call reduction: 60-70% through cache hits
  • Target cache hit rate: >70% within 100 ticks
  • Memory footprint: <100KB for typical usage
  • Added 37 comprehensive tests (25 unit, 12 regression)
  • Resolves issue ralphschuler/.screeps-gpt#1008 (implement path caching with TTL)
  • CPU Profiler Optimization: Reduced profiler overhead by 60-80% through caching and interval-based collection
  • Added tick-based caching for profiler state checks (isEnabledFast)
  • Reduced Memory.profiler.start access from 1000+ per tick to 1-2 per tick
  • Cache invalidated on profiler state changes (start/stop/clear) for correctness
  • Profiler overhead when stopped: reduced from ~1.0 CPU to ~0.3 CPU (70% reduction)
  • Profiler overhead when running: reduced by 30-40% due to caching optimization
  • Added DETAILED_STATS_INTERVAL (10 ticks) for StatsCollector expensive operations
  • Structure counts and construction sites collected every 10 ticks instead of every tick
  • Critical stats (CPU, creeps, energy) still collected every tick for monitoring
  • StatsCollector overhead: reduced from ~0.4-0.6 CPU to ~0.15-0.2 CPU (65% reduction)
  • Disabled diagnostic logging by default (enableDiagnostics: false) for production
  • Added runtime override via Memory.experimentalFeatures.statsDebug for debugging
  • Expected total CPU reduction: 2.0-2.5 CPU per tick (53-66% reduction)
  • Target baseline: <1.0 CPU per tick with 1 creep (from 3.77 CPU measured)
  • Resolves issue ralphschuler/.screeps-gpt#961 (optimize CPU profiler overhead)
  • CPU Profiler Optimization Runbook: Added comprehensive guide to docs/operations/runbooks.md
  • Documented build-time profiler disabling (PROFILER_ENABLED=false) for zero overhead
  • Added runtime profiler control guide (Profiler.start/stop/clear commands)
  • Added profiler data collection workflow and output interpretation
  • Defined performance baselines and CPU targets per creep count
  • Added optimization targets per component (Kernel, StatsCollector, BehaviorController)
  • Created regression test framework in tests/unit/profiler-caching.test.ts
  • Established performance thresholds for validation and regression detection
  • Emergency Spawn Bootstrap with Priority Refilling: Enhanced emergency spawn logic with spawn refilling priority for all creeps
  • CRITICAL: All creeps (harvesters, upgraders, builders) now prioritize spawn refilling above all other tasks when spawn energy drops below 50% capacity or below 150 energy (minimum spawn threshold)
  • Emergency creeps marked with emergency: true flag in memory for tracking and special handling
  • Added diagnostic message when emergency spawn cannot proceed due to insufficient energy (< 150 energy minimum)
  • Detects energy stuck in containers that cannot be transported without creeps
  • Logs energy availability, stored energy, and minimum threshold requirements
  • Enhanced emergency spawn success logs with energy percentage metrics
  • Visual indicators: 🚨spawn emoji and red pathfinding when creeps are in emergency spawn refill mode
  • Prevents spawn starvation by ensuring spawn always has minimum energy for continuous creep production
  • Universal application ensures both emergency recovery and normal operation maintain spawn energy reserves
  • Added comprehensive regression test suite (8 tests) for emergency deadlock scenarios
  • Helps identify and track recovery from spawn deadlock situations (0 creeps + low energy)
  • Resolves issue ralphschuler/.screeps-gpt#1002 (emergency spawn bootstrap implementation)
  • Parent issue: ralphschuler/.screeps-gpt#998 (zero creep population investigation)
  • Related: #959 (missing hauler role), #954 (storage automation), #688, #691 (energy validation)

• 0.104.0 – 2025-11-17 (changelog)

  • Optimized Creep Role Distribution at RCL4: Implemented dynamic role balancing to fix hauler spawning and workforce inefficiency
  • Added critical hauler priority system: spawns haulers first when storage/towers exist but hauler count is 0
  • Reduced harvester overstaffing from 4 to 2-3 when haulers are available for logistics
  • Implemented dynamic builder activation: scales 1-3 builders based on construction queue size (1-5 sites: 1, 6-15: 2, 16+: 3)
  • Implemented dynamic repairer activation: spawns repairers only when structures need maintenance
  • Added infrastructure detection: tracks construction sites and damaged structures for workforce optimization
  • Expected efficiency gain: ~30% improvement in energy throughput
  • Fixes all 3 failing regression tests in hauler-spawning-with-storage.test.ts
  • Resolves issue ralphschuler/.screeps-gpt#961 (optimize creep role distribution at RCL4)
  • Resolves issue ralphschuler/.screeps-gpt#955 (hauler role not spawning)
  • Profiler Data Collection Enablement: Added console backup mechanism to ensure profiler is always running
  • Created ensure-profiler-running.ts script that idempotently checks and starts profiler via console
  • Updated monitoring workflow with “Ensure profiler is running” step before data fetch
  • Adds resilience to existing auto-start feature (dual mechanism: auto-start on first tick + console backup every 30 min)
  • Created initial placeholder reports/profiler/latest.json to prevent health check failures
  • Added comprehensive unit tests for ensure-profiler-running script
  • Created profiler enablement verification documentation in docs/operations/profiler-enablement-verification.md
  • Enables function-level CPU profiling for proactive optimization and bottleneck analysis
  • Resolves issue ralphschuler/.screeps-gpt#856 (profiler data collection enablement)
  • Supports ralphschuler/.screeps-gpt#854 (profiler integration implementation)
  • Enables ralphschuler/.screeps-gpt#793 (CPU bucket-aware scheduler with profiler data)
  • Enables ralphschuler/.screeps-gpt#820 (performance baseline establishment)

• 0.89.5 – 2025-11-16 (changelog)

  • Context-Aware Build Validation Thresholds: Implemented context-aware MIN_SIZE validation to strengthen build safety
  • Monolithic main.js now requires 50KB minimum (ensures kernel + runtime components present)
  • Modular components maintain 500B minimum (preserves flexibility for type-only exports)
  • Validation leverages existing checkLoopExport parameter to distinguish file types
  • Added regression tests validating both threshold scenarios
  • Prevents broken bundler output from deploying to production
  • Resolves issue ralphschuler/.screeps-gpt#731 (context-aware MIN_SIZE validation)

• 0.83.23 – 2025-11-16 (changelog)

  • Task System Default State Inconsistency: Aligned build-time default with runtime default and documentation
  • Changed buildProject.ts default from "false" to "true" to match v0.32.0+ enabled-by-default design
  • Updated build configuration comments to document default values explicitly
  • Task system is now truly enabled by default at build time (matches runtime fallback behavior)
  • Set TASK_SYSTEM_ENABLED=false at build time to disable if needed
  • Added regression test to validate default task system state matches expected behavior
  • Resolves confusion between code comments claiming “enabled by default” and actual disabled behavior

• 0.83.7 – 2025-11-15 (changelog)

  • Memory.stats Collection Failure: Fixed TypeScript interface conflict preventing stats collection and causing monitoring blackout
  • Root cause: profiler/typings.d.ts declared interface Memory without declare global, creating conflicting local interface
  • This prevented the global Memory interface in types.d.ts from recognizing stats property
  • Solution: Removed conflicting Memory interface declaration from profiler typings
  • Added regression test memory-stats-interface.test.ts to prevent future interface conflicts
  • Resolves issue #684 (Memory.stats collection failure) and unblocks strategic monitoring capabilities
  • Restores PTR telemetry collection, enabling performance analysis and anomaly detection

• 0.83.5 – 2025-11-15 (changelog)

  • Overmind-RL Reinforcement Learning Research: Comprehensive evaluation of RL integration potential for bot AI optimization
  • Created research documentation in docs/research/overmind-rl-analysis.md
  • Analyzed Overmind-RL three-component architecture (Node.js backend, Python Gym wrapper, distributed training)
  • Evaluated RL algorithms (PPO, DQN), neural network designs, and reward function engineering
  • Assessed 7 use cases: combat micro, resource allocation, expansion, creep bodies, market trading, tasks, pathfinding
  • Documented training requirements: $2k-$10k first year, 870 hours effort, GPU infrastructure
  • Detailed cost-benefit analysis: RL 6x more expensive than proven Overmind patterns with uncertain ROI
  • Compatibility analysis: Architecture misalignment (Python vs. TypeScript-only), 10-200ms inference latency
  • Created 6-phase integration roadmap (33-48 weeks) if pursued in future
  • Decision: NOT RECOMMENDED for current integration—focus on proven optimization patterns instead
  • Defined revisit conditions: bot maturity (12-24 months), specific high-value use case, RL expertise, infrastructure budget
  • Updated TASKS.md with research findings and alternative recommendations
  • Related research: Overmind architecture (overmind-analysis.md), creep-tasks (#625), packrat (#626)

• 0.83.1 – 2025-11-15 (changelog)

  • Strategic Documentation Framework: Created comprehensive strategic documentation structure for tracking bot development phases and capturing learning insights
  • Created docs/strategy/ directory with phases/, learning/, and decisions/ subdirectories
  • Created strategic roadmap (docs/strategy/roadmap.md) documenting current phase status (Phase 1: 85%, Phase 2: 60%, Phase 3: 100%, Phase 4: 50%, Phase 5: 100%)
  • Documented success metrics, blockers (telemetry #791, container placement #783), and milestones
  • Created phase-specific documentation for all 5 development phases:
  • Phase 1: Foundation (RCL 1-2) - Bootstrap, basic economy, container harvesting
  • Phase 2: Core Framework (RCL 3-4) - Task system, spawn queue, link network
  • Phase 3: Advanced Economy (RCL 6-8) - Remote harvesting, terminal, labs, factory
  • Phase 4: Empire Coordination - Combat, traffic, expansion, multi-room logistics
  • Phase 5: Multi-Room & Global Management - Colony management, analytics, inter-shard communication
  • Migrated learning insights from CHANGELOG.md to structured documentation:
  • Bootstrap phase implementation pattern (v0.44.0) - Harvester-focused early-game optimization
  • Container-based harvesting pattern (v0.54.0) - Role specialization for efficiency
  • Round-robin task scheduling pattern (v0.57.1) - CPU fairness preventing creep starvation
  • Created Architectural Decision Records (ADR) template and guidelines
  • Updated AGENTS.md to reference strategic documentation in knowledge base
  • Updated README.md strategy section with links to roadmap, phases, learning insights, and ADRs
  • Enables strategic planning agent to analyze bot progression objectively
  • Captures institutional knowledge preventing repeated failed approaches
  • Provides context for autonomous agents making improvement decisions
  • Overmind Architecture Research: Comprehensive analysis of Overmind bot patterns for potential integration
  • Created research documentation in docs/research/overmind-analysis.md
  • Identified 12 key architectural patterns with compatibility assessments
  • Documented quick wins: Task Persistence, Decorator-Based Caching, Directive System
  • Prioritized implementation roadmap with complexity and value estimates
  • Linked patterns to existing issues (#478, #487, #494, #392, #426, #493, #607, #614)
  • Added recommendations to TASKS.md for phased implementation
  • Total of 10 prioritized patterns spanning Phases 2-5
  • Focused on high-value areas: CPU optimization, task management, multi-room scaling

• 0.80.0 – 2025-11-15 (changelog)

  • CI Autofix Circuit Breaker: Implemented circuit breaker to prevent infinite retry loops in CI autofix workflow
  • Added circuit breaker logic to track consecutive failures (max 3 attempts)
  • Implemented 15-minute backoff period between retry attempts
  • Created automatic escalation to GitHub issues when circuit breaker trips
  • Added diagnostic logging showing retry attempts and circuit breaker status
  • Circuit breaker resets on successful autofix or after backoff period
  • Prevents workflow saturation from repeated failed autofix attempts
  • Checks for existing escalation issues to avoid duplicates
  • Added comprehensive regression test suite (16 tests) in tests/regression/ci-autofix-circuit-breaker.test.ts
  • Resolves issue: fix(automation): CI autofix workflow saturation - 17+ consecutive action_required failures

• 0.76.0 – 2025-11-14 (changelog)

  • Dying Creep Energy Dropping: Implemented automatic energy dropping behavior for creeps approaching end of life
  • Creeps with TTL below threshold (default: 50 ticks) automatically drop all carried energy
  • Prevents energy waste from creep despawning
  • Configurable via Memory.dyingCreepBehavior (enabled/disabled and threshold)
  • Visual feedback: dying creeps display “💀” emoji
  • Works with both task-based and role-based execution systems
  • Minimal CPU overhead: ~0.01 CPU per creep per tick
  • Created helper functions in packages/bot/src/runtime/behavior/creepHelpers.ts:
  • isCreepDying(creep, threshold): Detects dying creeps
  • handleDyingCreepEnergyDrop(creep): Handles energy drop logic
  • Integrated into BehaviorController pre-execution checks
  • Added 11 unit tests in tests/unit/creepHelpers.test.ts
  • Added 8 integration tests in tests/unit/behaviorController.test.ts
  • Created comprehensive documentation in docs/runtime/creep-lifecycle.md
  • Total test coverage: 795 passing tests
  • Resolves issue: feat(runtime): implement energy dropping behavior for dying creeps

• 0.64.0 – 2025-11-13 (changelog)

  • Build Validation Enhancements: Implemented comprehensive build artifact validation to prevent corrupted or empty files from being deployed
  • Added file size validation (non-zero check and 500-byte minimum threshold)
  • Added content validation to verify main.js exports the required loop function
  • Enhanced validation for both monolithic and modular build architectures
  • Validation applies to all module files in modular builds (behavior.js, bootstrap.js, etc.)
  • Exported validateFile() function for testing and reusability
  • Created 12 comprehensive unit tests in tests/unit/build-validation.test.ts
  • Tests cover empty files, small files, missing exports, and various export patterns
  • Updated packages/utilities/scripts/lib/buildProject.ts with enhanced validation logic
  • Prevents critical deployment failures by catching build errors early in the pipeline
  • Resolves issue: fix(build): enhance build validation to check file size and content validity
  • Strategic Planning Automation: Implemented autonomous strategic planning agent that analyzes bot performance and creates improvement roadmaps
  • Added workflow: .github/workflows/copilot-strategic-planner.yml (runs every 8 hours)
  • Created comprehensive strategic planner prompt: .github/copilot/prompts/strategic-planner
  • Integrates bot snapshots, PTR telemetry, profiler data, and documentation for strategic analysis
  • Identifies improvement opportunities across six categories: performance, economy, expansion, defense, infrastructure, automation
  • Creates evidence-based issues with clear priorities, acceptance criteria, and implementation approaches
  • Updates strategic documentation to maintain alignment with bot capabilities
  • Learning feedback loop from past implementations to avoid repeating mistakes
  • Detailed documentation in docs/automation/strategic-planning.md
  • Updated docs/automation/overview.md with strategic planning workflow description
  • Completes autonomous development loop: Monitoring → Strategic Planning → Implementation → Validation
  • Resolves issue: feat(automation): implement strategic planning Copilot agent for autonomous bot improvement

• 0.57.1 – 2025-11-12 (changelog)

  • Task System CPU Starvation Prevention: Implemented round-robin scheduling to ensure fair creep execution under CPU constraints
  • Added tickOffset rotation to prevent same creeps from being consistently skipped
  • Added lastExecuted tracking map to monitor execution gaps per creep
  • Added getStarvationStats() method for monitoring fairness metrics
  • With 25 creeps and CPU allowing 12/tick: old system permanently starved 13 creeps, new system cycles all with max 14-tick gaps
  • Created 13 unit tests in tests/unit/taskManager-round-robin.test.ts validating fair scheduling
  • Created 7 regression tests in tests/regression/task-system-cpu-starvation.test.ts for high creep count scenarios
  • Updated docs/runtime/task-system.md with round-robin scheduling documentation
  • All creeps now get equal opportunity to execute tasks, eliminating permanent starvation
  • Resolves issue: Task system CPU threshold checking may cause starvation with high creep counts
  • StatsCollector Error Handling: Added comprehensive error handling to prevent silent stats collection failures
  • Wrapped main collection in try/catch to prevent exceptions from blocking Memory.stats writes
  • Isolated room stats collection to prevent one bad room from breaking entire collection
  • Added fallback mechanism to ensure Memory.stats always has valid structure with safe property access
  • Resolves #658: Bot executes normally but Stats API returns empty data

• 0.54.0 – 2025-11-12 (changelog)

  • Container-Based Harvesting Automation: Implemented dynamic role adjustment system that transitions to efficient container-based economy when infrastructure is ready
  • Added repairer role for structure maintenance (prioritizes roads and containers, then other structures)
  • Repairer body optimized for repair work: 2 WORK, 1 CARRY, 2 MOVE (300 energy cost)
  • System automatically detects containers near energy sources and adjusts role spawning:
  • Spawns 1 stationary harvester per source with adjacent container
  • Spawns 2 haulers per controlled room for energy transport
  • Spawns 1 repairer per controlled room for infrastructure maintenance
  • Reduces regular harvester minimum from 4 to 2 when using container-based system
  • Added repairer memory interface and task constants (repairerGather, repair)
  • Created comprehensive test suite in tests/unit/repairer.test.ts (3 tests)
  • Repairer gathers energy from containers/storage, repairs infrastructure prioritizing roads/containers
  • System seamlessly transitions between mobile harvesters and stationary+hauler economy
  • Resolves #667: Add repairer and hauler to the system with container-based automation

• 0.52.0 – 2025-11-12 (changelog)

  • Bot Aliveness Heartbeat Monitoring: Implemented automated bot heartbeat monitoring with graduated failure detection to catch outages within 30 minutes instead of 8+ hours
  • Created packages/utilities/scripts/check-bot-health.ts with multi-stage health checks (PTR stats → world-status API → console fallback)
  • Implemented graduated alert thresholds: 0-15 min (silent), 15-30 min (warning), 30-60 min (HIGH), 60+ min (CRITICAL)
  • Added persistent health state tracking in reports/monitoring/health.json with 100-entry detection history
  • Integrated health check into .github/workflows/screeps-monitoring.yml (runs every 30 minutes)
  • Enhanced packages/utilities/scripts/check-ptr-alerts.ts to generate bot outage alerts with push/email notifications
  • Added comprehensive test suite tests/unit/check-bot-health.test.ts with 9 tests covering persistence, thresholds, and history tracking
  • Updated documentation in docs/automation/autonomous-monitoring.md with Phase 3.5: Bot Aliveness Heartbeat specification
  • Health state committed alongside bot snapshots for historical tracking and trend analysis
  • Addresses monitoring blind spot where bot death was indistinguishable from telemetry failures
  • Resolves #561: Implement automated bot aliveness heartbeat with early failure detection

• 0.51.7 – 2025-11-12 (changelog)

  • Spec-Kit Workflow System: Removed unused specification-driven development workflow infrastructure
  • Deleted workflow file .github/workflows/copilot-speckit.yml
  • Deleted prompt templates .github/copilot/prompts/speckit-plan and speckit-refine
  • Deleted documentation docs/automation/spec-kit-workflow.md from all documentation locations
  • Removed speckit label definition from .github/labels.yml
  • Removed all spec-kit references from automation documentation
  • Deleted regression test tests/regression/speckit-workflow-structure.test.js
  • Repository has standardized on Todo automation workflow for implementations
  • Reduces maintenance burden and simplifies automation surface area
  • Resolves #557: Remove unused spec-kit workflow and documentation

• 0.51.3 – 2025-11-12 (changelog)

  • Deprecation Strategy and Code Lifecycle Management: Implemented comprehensive system for managing deprecated code and technical debt
  • Created deprecation policy document in docs/development/deprecation-policy.md with lifecycle phases and guidelines
  • Created deprecation registry in docs/development/deprecation-registry.md to track all active deprecations
  • Created migration guide template in docs/development/migration-guide-template.md for consistent documentation
  • Added ESLint rule @typescript-eslint/no-deprecated to warn when deprecated APIs are used
  • Added CI workflow .github/workflows/guard-deprecation.yml for automated deprecation checks
  • Documented current deprecations: role-based behavior system and legacy label system
  • Established 2-3 release cycle deprecation timeline for minor versions
  • Integrated deprecation tracking with CHANGELOG format
  • Provides clear upgrade paths and reduces technical debt accumulation
  • Resolves #556: Implement deprecation strategy and code lifecycle management system

• 0.51.1 – 2025-11-12 (changelog)

  • PTR Telemetry Blackout Regression: Implemented comprehensive prevention measures for recurring stats collection failures (#550)
  • Added validation to StatsCollector to detect Memory.stats write failures
  • Created scripts/validate-telemetry-health.ts for automated health checks
  • Integrated telemetry health validation into monitoring workflow (runs every 30 minutes)
  • Added post-deployment validation step (5 min wait + health check)
  • Created regression test suite tests/regression/stats-collection-blackout.test.ts with 8 test cases
  • Enhanced documentation in docs/operations/stats-collection.md with troubleshooting and recovery procedures
  • Implements automated detection of empty stats within 15 minutes
  • Prevents recurrence of issues #523, #331, #345 through proactive monitoring
  • Resolves #550: PTR telemetry blackout regression - empty stats data despite successful deployments

• 0.50.0 – 2025-11-12 (changelog)

  • Builder Wall Maintenance: Builder role now repairs walls and ramparts to configurable target HP thresholds
  • Added WALL_TARGET_HP constant (100,000 HP) for wall repair threshold
  • Added RAMPART_TARGET_HP constant (50,000 HP) for rampart repair threshold
  • Modified builder repair filter to include walls/ramparts below target HP
  • Fixes #644: Builder role now maintains defensive structures

• 0.48.0 – 2025-11-12 (changelog)

  • Issue Triage Enhancement: Enhanced issue triage automation with comprehensive context gathering
  • Added code context gathering using GitHub MCP search_code to find related files in .github/, src/, tests/
  • Added issue cross-referencing to identify related open and closed issues
  • Added PR cross-referencing to find related pull requests
  • Enhanced reformulated issue body with new sections: “Related Code”, “Related PRs”
  • Updated triage comment to include discovered context
  • Extended JSON output to include related_prs and related_code_files fields
  • Resolves #639: Make issue triage automation check current code, issues, and pull requests

• 0.47.1 – 2025-11-11 (changelog)

  • Monorepo Restructuring: Reorganized repository into packages-based monorepo structure
  • Created /packages directory with bot, docs, utilities, actions, and console packages
  • Migrated src/ to packages/bot/src/ for core Screeps AI implementation
  • Migrated scripts/ to packages/utilities/scripts/ for build tooling and utilities
  • Migrated docs/, source/, themes/, _config.yml to packages/docs/ for documentation site
  • Configured Bun workspaces in root package.json for monorepo dependency management
  • Updated TypeScript path aliases to reference new package locations
  • Updated ESLint configuration to lint new package structure
  • Updated vitest configuration for new source paths
  • Updated all GitHub workflows to reference new package paths
  • Updated all import paths in tests and utilities
  • All 580 tests passing, lint passing, build working
  • Improved code organization with clear package boundaries
  • Enables independent package versioning and deployment
  • Better separation of concerns across codebase
  • Resolves #[issue-number]: Restructure repository into monorepo with packages organization

• 0.44.0 – 2025-11-10 (changelog)

  • Bootstrap Phase: Implemented automated first-room resource optimization with harvester-focused spawning
  • Added BootstrapPhaseManager class for bootstrap phase state management
  • Integrated bootstrap logic with Kernel and BehaviorController
  • Adjusts role minimums during bootstrap phase (6 harvesters, 1 upgrader, 0 builders = 80%+ harvesters)
  • Automatically activates for new rooms with controller level < 2
  • Exits when controller level 2 reached OR stable infrastructure (4+ harvesters, 300+ energy)
  • Tracks bootstrap state in Memory with persistence across code reloads
  • Configurable completion criteria via BootstrapConfig
  • Comprehensive documentation in docs/runtime/bootstrap.md
  • 37 unit tests validating bootstrap activation, completion, role minimums, and integration
  • Resolves #530: Implement bootstrap phase for optimal first-room resource utilization
  • Operational Milestones Documentation: Created docs/operations/milestones.md for tracking progression achievements
  • Documented E46S58 controller level 2 upgrade milestone (2025-11-08, shard3)
  • Established milestone tracking framework for controller upgrades, territorial expansion, and infrastructure development
  • Includes automation recommendations for future milestone detection
  • Related to #533: Monitoring verification for controller upgrade detection

• 0.43.1 – 2025-11-10 (changelog)

  • Console Telemetry Fallback: Fixed “expression size too large” error by implementing chunked query strategy
  • Split single large console command into 5 smaller, focused queries (CPU, GCL, rooms, creeps, resources)
  • Each query limited to 1000-1200 characters to stay within Screeps API limits
  • Added retry logic with exponential backoff (3 attempts, 1s/2s/4s delays)
  • Added expression size validation before sending commands to API
  • Restored monitoring resilience when Stats API returns empty data
  • Added comprehensive test suite (tests/unit/fetch-console-telemetry.test.ts) with 9 test cases
  • Resolves #526: Console fallback “expression size too large” error
  • Related to #523: PTR telemetry blackout requiring console fallback

• 0.40.0 – 2025-11-10 (changelog)

  • Pathfinding Abstraction Layer: Integrated screeps-cartographer for advanced pathfinding optimization
  • Created PathfindingProvider interface for flexible pathfinding implementations
  • Implemented DefaultPathfinder using native Screeps PathFinder (baseline)
  • Implemented CartographerPathfinder using screeps-cartographer library
  • Added PathfindingManager with configuration system to toggle between providers
  • Integrated with TaskManager and TaskAction for task-based movement
  • Added pathfindingProvider configuration option to BehaviorController and TaskManager
  • Maintains backward compatibility with native pathfinding as default
  • Comprehensive documentation in docs/runtime/pathfinding.md
  • 14 unit tests covering provider selection, configuration, and behavior
  • Build size increased from 579.6kb to 713.2kb (+134kb for screeps-cartographer)
  • No security vulnerabilities detected in new dependency
  • Provides foundation for CPU-efficient pathfinding with caching and optimization
  • Task system automatically uses configured pathfinding provider
  • Test infrastructure updated to properly mock screeps-cartographer (Game, Memory, PathFinder globals)
  • All 506 existing tests continue to pass
  • Addresses #533: screeps-cartographer integration for advanced pathfinding

• 0.39.1 – 2025-11-10 (changelog)

  • Console Output TypeError: Fixed “Cannot convert object to primitive value” error in MemoryValidator
  • Changed from result.error.message to JSON.stringify(result.error.issues) for proper Zod error serialization
  • Zod error objects lack a simple message property, causing primitive conversion errors in Screeps console
  • Added comprehensive regression test suite (tests/regression/console-output-type-error.test.ts) with 6 test cases
  • Ensures all console logging properly handles complex objects without type conversion errors
  • Resolves console errors reported via email from noreply@screeps.com (2025-11-07)

• 0.37.3 – 2025-11-10 (changelog)

  • Jon Winsley Screeps Blog Analysis: Comprehensive review of 29 blog posts documenting real-world Screeps development patterns
  • Created detailed analysis document at docs/strategy/external-analysis/jon-winsley-analysis.md
  • Extracted key insights on CPU optimization, task management, memory serialization, and strategic planning
  • Identified 50+ actionable implementation items organized by priority
  • Cross-referenced patterns with existing repository issues (#392, #426, #478, #487, #493, #494, #495, #496, #468)
  • Documented “Great Purge” philosophy: simplicity over complexity for better performance
  • Analyzed Objective-based architecture as alternative to centralized task systems
  • Extracted monitoring and metrics patterns for PTR telemetry enhancement
  • Documented remote mining logistics and multi-room scaling strategies
  • Added comprehensive action items to TASKS.md for systematic implementation
  • Complements existing Screeps Quorum analysis with individual developer perspective

• 0.33.5 – 2025-11-09 (changelog)

  • Account Upgraded to Lifetime Subscription: Screeps account upgraded from free tier to lifetime subscription
  • CPU allocation increased from 20 to 50 (150% increase, +30 CPU)
  • Memory allocation remains at 2048 KB (unchanged)
  • Updated CPU thresholds to reflect new resource allocation:
  • PerformanceTracker: highCpuThreshold 70% → 75%, criticalCpuThreshold remains 90%
  • SystemEvaluator: cpuUsageWarningRatio 80% → 85%, cpuCriticalRatio remains 95%
  • BehaviorController: cpuSafetyMargin 80% → 85%
  • Kernel: cpuEmergencyThreshold remains 90% (last line of defense)
  • Benefits: Reduced CPU constraint concerns, enables advanced features (profiler, task management system) with better margins, reduces timeout risk
  • Strategic opportunity: Focus shifts from aggressive CPU optimization to sophisticated AI development with better safety margins
  • Documentation: Added comprehensive resource allocation guide at docs/operations/resource-allocation.md
  • Impact: Helps mitigate CPU timeout incidents (#468, #494) and enables evaluation of previously disabled features (#478, #475)

• 0.31.11 – 2025-11-09 (changelog)

  • Removed Unnecessary Credential Exposure in Monitoring Workflow: Eliminated SCREEPS_EMAIL and SCREEPS_PASSWORD from screeps-monitoring.yml workflow
  • Credentials were unnecessarily exposed to GitHub Copilot CLI during monitoring operations
  • Scripts (fetch-profiler-console.ts, check-ptr-alerts.ts) use only SCREEPS_TOKEN for API authentication
  • Follows least-privilege principle by providing only token-based read-only access
  • Reduces attack surface by removing write-capable credentials from third-party AI service environment
  • Aligns with deploy.yml workflow pattern which demonstrates token-only authentication is sufficient
  • Impact: High priority security improvement - eliminated credential exposure in workflow running every 30 minutes
  • No functionality changes - all monitoring features (PTR stats, profiler fetch, alert checking) continue to work with token-only authentication

• 0.31.9 – 2025-11-09 (changelog)

  • Enhanced TypeScript Type Safety with Runtime Validation: Improved type safety throughout the codebase to prevent runtime errors
  • Added lastTimeoutTick field to Memory.stats interface in types.d.ts for CPU timeout detection tracking
  • Created MemoryValidator class (src/runtime/memory/MemoryValidator.ts) using zod schemas for runtime validation of Memory structures (infrastructure code only, not yet integrated into runtime)
  • Implemented validateStats() method to validate Memory.stats against TypeScript interface definitions (available for future use)
  • Implemented validateAndRepairStats() method to automatically repair corrupted memory with sensible defaults (available for future use)
  • Added validateGameContext() function in src/main.ts to replace unsafe Game as unknown as GameContext type casting with explicit runtime checks
  • Enhanced error handling with specific error classification (TypeError vs Error vs unknown errors)
  • Added 26 unit tests for memory validation infrastructure and error classification (note: validateGameContext tests validate error types but not the validation function itself due to it being private)
  • Strengthened ESLint configuration with stricter TypeScript rules for runtime files (src/runtime/**/*.ts):
  • @typescript-eslint/no-explicit-any: error (prevents unsafe any types)
  • @typescript-eslint/prefer-nullish-coalescing: error (safer null/undefined handling)
  • @typescript-eslint/prefer-optional-chain: error (cleaner optional property access)
  • @typescript-eslint/no-unsafe-assignment: error (prevents unsafe type assignments in runtime code)
  • @typescript-eslint/no-unsafe-call: error (prevents calls without proper type checking in runtime code)
  • @typescript-eslint/no-unsafe-member-access: error (prevents unsafe property access in runtime code)
  • Exception added for src/main.ts to allow profiler initialization with unsafe operations required for global exposure
  • All tests pass (477 unit tests), build succeeds
  • Improves autonomous monitoring compatibility with parseable error messages
  • Benefits CPU timeout detection, evaluation system, and autonomous improvement workflows

• 0.31.5 – 2025-11-08 (changelog)

  • Security Audit Workflow jq Parsing Error: Fixed shell parsing error in .github/workflows/guard-security-audit.yml that was causing workflow failures
  • Root cause: Multi-line jq command with backslash continuation was causing syntax errors in shell execution
  • Fixed by properly formatting multi-line jq command without trailing backslashes before pipe operators
  • Added defensive error handling for jq parsing failures to prevent workflow crashes
  • Added handling for “unknown” vulnerability counts when parsing fails
  • Created regression test (tests/regression/guard-security-audit-workflow-syntax.test.ts) to validate workflow syntax and error handling
  • Workflow now passes yamllint validation with 80-character line limit
  • Fixes consistently failing Security Audit workflow (run 19183827409 and subsequent runs)
  • Restores CI/CD pipeline health for security vulnerability scanning

• 0.31.1 – 2025-11-08 (changelog)

  • Screeps Quorum Architecture Analysis: Comprehensive analysis of the Screeps Quorum community-driven bot architecture
  • Created docs/strategy/external-analysis/screeps-quorum-analysis.md with detailed architectural pattern review
  • Analyzed community governance automation (GitConsensus), deployment architecture (CircleCI + Gulp), modular code organization, and monitoring patterns (ScreepsDashboard)
  • Identified high-priority recommendations: Multi-agent consensus protocol, runtime version tracking, QoS monitoring system
  • Compared Screeps GPT architecture with Screeps Quorum to validate current design decisions
  • Documented actionable recommendations with implementation roadmap aligned to Phase 1-2 deliverables
  • Analysis supports strategic planning (Issue #23), specialized GitHub Actions (Issue #210), and enhanced Copilot workflows (Issue #89)
  • Documented Zero Security Vulnerabilities in Dependency Tree (#288)
  • Comprehensive audit confirmed all previous vulnerabilities (from #125) remain resolved
  • npm audit reports 0 vulnerabilities across all severity levels (critical, high, moderate, low)
  • Validated that axios security fixes (axios@1.13.2) from Screeps GPT release 0.19.3 (2025-11-07) are still effective
  • Security audit workflow (guard-security-audit.yml) confirmed operational with daily scheduled runs
  • Issue #288 determined to be duplicate of already-resolved #125 (closed 2025-11-07)
  • All 451 unit tests passing, build successful, no security blockers for deployment
  • Monitoring system data appears to have been based on stale/cached vulnerability information

• 0.29.3 – 2025-11-08 (changelog)

  • Repository Cleanup and Standardization: Comprehensive cleanup to improve maintainability and consistency
  • Removed unnecessary documentation files: DEPLOY_WORKFLOW_FIX.md, HEXO_IMPLEMENTATION_NOTES.md, IMPLEMENTATION.md, IMPLEMENTATION_LOG.md
  • Removed validation script: validate-deploy-fix.sh
  • Removed backup file: package.json.backup
  • Removed deprecated ESLint configuration files: .eslintrc.cjs, .eslintrc-polyfill.cjs
  • Removed structuredClone polyfill (no longer needed with Node.js 18+ requirement)
  • Reorganized build system: moved scripts/buildProject.ts to scripts/lib/buildProject.ts
  • Updated .gitignore to prevent future similar issues (backup files, temporary docs, implementation notes)
  • Updated ESLint flat config to remove polyfill dependency (Node.js 18+ has native structuredClone)
  • Updated regression tests to reflect Node.js 18+ requirement and removal of polyfill
  • All existing functionality preserved (builds, tests, deployments work unchanged)

• 0.24.0 – 2025-11-08 (changelog)

  • Self-Healing Memory System: Automatic detection and repair of corrupted memory structures
  • New MemorySelfHealer class validates and repairs core memory structures (creeps, rooms, roles, respawn, stats, systemReport)
  • Detects circular references, invalid types, malformed data, and missing structures
  • Automatic repair of corrupted entries with configurable auto-repair behavior
  • Emergency reset capability for complete memory corruption
  • Integrated into Kernel bootstrap (runs before migrations and other operations)
  • Enabled by default (enableSelfHealing: true in Kernel config)
  • 28 comprehensive unit tests covering validation, repair, and emergency reset scenarios
  • Updated documentation with self-healing usage, best practices, and troubleshooting guide

• 0.19.3 – 2025-11-07 (changelog)

  • Resolved 3 high-severity vulnerabilities in axios dependency (#282)
  • Fixed GHSA-jr5f-v2jv-69x6: axios SSRF and credential leakage vulnerability (CVE affecting axios < 0.30.0)
  • Fixed GHSA-4hjh-wcwx-xvwj: axios DoS attack through lack of data size check (CVE affecting axios < 0.30.2)
  • Applied npm package overrides to force axios@1.13.2 across all transitive dependencies
  • Transitive dependency path: screeps-api@1.16.1 → axios@0.28.1 (vulnerable) → axios@1.13.2 (patched)
  • All tests, builds, and linting pass with no regressions

• 0.18.0 – 2025-11-07 (changelog)

  • Phase 3 Advanced Economy: Complete implementation of RCL 6-8 economy features
  • LinkManager: Automated energy distribution through link networks with role-based classification
  • TerminalManager: Inter-room resource logistics with priority-based transfer queue
  • LabManager: Compound production and creep boosting with state management (Tier 1 compounds only)
  • FactoryManager: Automated commodity production with priority queue system
  • Phase 4 Empire Coordination: Initial implementation of empire-wide coordination features
  • CombatManager: Squad-based combat coordination with threat assessment
  • TrafficManager: Priority-based movement coordination with collision avoidance
  • Unit tests for all new managers (LinkManager, TerminalManager, LabManager, FactoryManager, CombatManager, TrafficManager)
  • Documentation for Phase 3 and Phase 4 features in docs/automation/overview.md
  • Optional Memory persistence support for all Phase 3/4 managers with RoomPosition serialization
  • Fixed RoomPosition serialization in LinkManager and TrafficManager Memory persistence
  • Fixed recipe system in LabManager to use Screeps constants instead of string literals
  • Fixed engagement double-counting in CombatManager for hybrid attack/ranged creeps
  • Fixed squad ID generation to prevent collisions using member composition hash
  • Fixed boost array mutation during iteration in LabManager
  • Optimized TerminalManager to avoid redundant room.find() calls

• 0.14.0 – 2025-11-07 (changelog)

  • Modular Build Documentation and Validation (#506)
  • Updated modular deployment documentation to accurately reflect ES2018 target (was incorrectly documented as ES2021)
  • Expanded module list documentation to include all 15 generated modules (behavior, bootstrap, defense, evaluation, infrastructure, memory, metrics, planning, respawn, scouting, tasks, types, utils, visuals)
  • Updated deployment size estimates to reflect actual builds (~95KB single bundle vs ~384KB modular)
  • Added build-time validation to verify expected artifacts are generated
  • Validation ensures main.js is always present and all runtime modules are generated in modular builds
  • ES2018 target compliance regression test now passes with complete modular build validation
  • Root cause: Documentation was outdated and lacked validation to catch missing artifacts during builds

• 0.12.0 – 2025-11-06 (changelog)

  • Comprehensive Memory Management and Migration System (#490)
  • Implemented MemoryGarbageCollector for automated cleanup of stale data
  • Removes orphaned room data from inactive/abandoned rooms with configurable retention (default: 10000 ticks)
  • Rotates old system evaluation reports to prevent memory bloat (default: 500 ticks retention)
  • Incremental cleanup with CPU throttling via maxCleanupPerTick parameter (default: 10 items/tick)
  • Runs every 10 ticks by default, can be disabled via kernel config
  • Implemented MemoryMigrationManager for schema versioning and safe migrations
  • Adds Memory.version field for tracking schema version
  • Migration registry system for version-specific upgrade handlers
  • Automated migration execution on version changes with error handling
  • Memory integrity validation after migrations
  • Built-in migration for version tracking initialization
  • Implemented MemoryUtilizationMonitor for memory usage tracking and overflow prevention
  • Real-time memory usage measurement with per-subsystem breakdown
  • Configurable warning (default: 70%) and critical (default: 90%) thresholds
  • Allocation capacity checks via canAllocate() method
  • Memory budgeting system for subsystems via getBudget() method
  • Human-readable byte formatting for monitoring output
  • Integrated memory management into Kernel orchestration
  • Migrations run automatically on version changes
  • Garbage collection runs every 10 ticks (configurable)
  • Memory utilization measured and passed to evaluation system
  • Enhanced SystemEvaluator with memory health monitoring
  • Memory utilization findings with warning/critical severity levels
  • Reports largest memory-consuming subsystems in critical alerts
  • Recommendations for garbage collection and retention tuning
  • Added comprehensive test coverage: 22 new unit tests covering all components
  • Created detailed documentation in docs/runtime/memory-management.md
  • Architecture overview and component descriptions
  • Configuration options and usage examples
  • Integration patterns with kernel and evaluator
  • Best practices for GC, migrations, and optimization
  • Troubleshooting guide and performance considerations
  • Memory management system designed for minimal CPU overhead (~0.5-2 CPU per GC cycle)
  • Supports configurable retention policies for different data types
  • All 231 unit tests pass, build successful, linter clean

• 0.11.3 – 2025-11-06 (changelog)

  • Resolved ReferenceError: process is not defined in Screeps runtime (#488)
  • Updated esbuild configuration in scripts/buildProject.ts to define process.env.TASK_SYSTEM_ENABLED and process.env.ROOM_VISUALS_ENABLED at build time
  • Environment variables are now replaced with literal values during bundling, eliminating runtime dependency on Node.js process global
  • Added regression test tests/regression/nodejs-globals-bundle.test.ts to prevent Node.js global contamination in future builds
  • Validates that bundled code contains no process.env, process object access, require() calls, __dirname, or __filename references
  • Fixes critical runtime error that blocked all bot execution in Screeps sandbox environment
  • Environment variables default to "false" if not set, can be enabled by setting TASK_SYSTEM_ENABLED=true or ROOM_VISUALS_ENABLED=true during build
  • Memory-based feature flags (Memory.experimentalFeatures) remain functional in Screeps runtime

• 0.8.0 – 2025-11-06 (changelog)

  • Migrated pre-commit hook to use ESLint flat config and removed 34 unused eslint-disable directives
  • Updated .husky/pre-commit to use ESLint flat config (eslint.config.mjs) instead of deprecated .eslintrc.cjs
  • Removed export ESLINT_USE_FLAT_CONFIG=false from pre-commit hook to enable flat config usage
  • Removed 34 unused eslint-disable directives across 7 files (3 scripts, 4 test files)
  • Scripts: fetch-console-telemetry.ts, fetch-profiler-console.ts, fetch-resilient-telemetry.ts
  • Test files: resilient-monitoring.test.ts, basePlanner.test.ts, constructionManager.test.ts, profiler.test.ts
  • Directives were marked as unused by flat config which doesn’t apply strict type-checking rules to tests/scripts
  • Old .eslintrc.cjs config applied strict rules everywhere, making directives appear necessary
  • Flat config (.eslintrc.mjs) applies type-checking rules only to src/**/*.ts for better performance
  • Improves code readability and maintainability by removing unnecessary ESLint pragmas
  • All tests pass after cleanup, confirming no functional impact
  • Addresses consistency between lint configuration and pre-commit validation
  • Workflow concurrency controls for race condition prevention
  • Added concurrency controls to 20 GitHub Actions workflows that previously lacked them
  • Guard workflows (10 files): Use per-workflow-ref grouping with cancel-in-progress: true to optimize PR validation
  • Copilot workflows (7 files): Use workflow-specific concurrency groups with cancel-in-progress: false to prevent agent conflicts
  • Monitoring workflows (2 files): Use workflow-level concurrency with cancel-in-progress: false to prevent overlapping runs
  • Singleton workflows (2 files): Use workflow-level concurrency with cancel-in-progress: true for safety
  • Created comprehensive regression test tests/regression/workflow-concurrency.test.ts to validate all workflows have proper concurrency controls
  • Prevents race conditions in guard workflow runs when multiple commits are pushed rapidly
  • Eliminates resource waste from stale workflow runs continuing after force-push
  • Prevents multiple Copilot agents from working on the same issue simultaneously
  • Prevents monitoring workflow overlaps that could cause API rate limit issues
  • Addresses ralphschuler/.screeps-gpt#50 and provides detailed implementation for concurrency control requirements
  • Enhanced pre-commit hook with regression and coverage tests
  • Added test:regression to pre-commit hook to enforce regression test execution before commits
  • Added test:coverage to pre-commit hook to enforce coverage checks before commits
  • Ensures all quality gates are enforced for agent commits, preventing regressions and coverage drops
  • Addresses issue #469 - Make sure to include the regression and coverage tests into the pre-commit hook
  • Regression test alignment with workflow versions
  • Updated tests/regression/post-merge-workflow-git-race-condition.test.ts to expect git-auto-commit-action@v7 instead of v5
  • Aligns test expectations with actual workflow configuration in .github/workflows/post-merge-release.yml
  • Test isolation for build-dependent regression tests
  • Made modular-build.test.ts and es2018-target.test.ts run sequentially to prevent race conditions
  • Added file-based locking mechanism to prevent parallel execution across test files
  • Prevents parallel test execution from interfering with shared dist directory during builds
  • Uses exponential backoff for lock acquisition to handle race conditions gracefully
  • Phase 1 Foundation Features
  • Added PixelGenerator class for automatic pixel generation when CPU bucket is full (default: 10000)
  • Implemented Logger utility with structured logging, timestamps, and log levels (debug, info, warn, error)
  • Added child logger support for contextual logging
  • Created unit tests for PixelGenerator validating deterministic output and bucket threshold logic
  • Created unit tests for Logger validating log levels, timestamps, context, and child logger functionality
  • Created unit tests for memory bootstrapping validating memory initialization and reset behavior
  • Created regression test for extension placement at RCL 1-2 validating bunker layout pattern
  • Integrated PixelGenerator into Kernel for automatic pixel generation on each tick
  • All new features include TSDoc comments for exported functions and classes
  • Addresses Phase 1 requirements: pixel generation, structured logging, memory tests, extension placement tests
  • Codex composite action for automation workflows
  • Added .github/actions/codex-exec as a drop-in replacement for copilot-exec
  • Wraps the official openai/codex-action@v1 with repository-specific caching and prompt rendering
  • Maintains existing inputs/outputs while renaming the token parameter to codex-token for clarity
  • CPU timeout incident tracking documentation
  • Created centralized incident tracking document in docs/operations/cpu-timeout-incidents.md
  • Documents systematic CPU timeout pattern on shard3 (6 incidents spanning 2025-10-26 to 2025-10-27)
  • Added Incident #6 (2025-10-27 03:38 UTC) at main:872:22 continuing systematic timeout pattern
  • Updated temporal pattern analysis showing pattern continuation across two days
  • Updated systematic indicators showing ongoing degradation despite CPU optimization efforts
  • Integrates with existing systematic resolution framework (#396, #417, #380, #391)
  • References architectural prevention solutions (#364, #392, #299)
  • Documents infrastructure dependencies (#428, #420 - PTR telemetry blackout)
  • Provides coordination approach for systematic resolution vs individual tactical fixes
  • Tracks incident patterns, temporal distribution, and location analysis
  • Documents monitoring integration with PTR telemetry and runtime evaluation
  • Establishes acceptance criteria for incident documentation and systematic resolution
  • Addresses Incident #6 - CPU timeout at main:872:22 on shard3 requiring systematic coordination
  • Enhanced copilot-ci-autofix workflow effectiveness
  • Added comprehensive failure classification system with 6 automatic fix categories and 5 manual review categories
  • Improved error context gathering with full log downloads, error indicator extraction, and related failure analysis
  • Implemented specialized fix strategies for each failure type (linting, formatting, version sync, dependencies, documentation, compilation)
  • Enhanced workflow configuration with 45-minute timeout and verbose logging for better debugging
  • Expanded JSON output format with failure_type, fix_strategy, validation_commands, and files_changed fields for metrics tracking
  • Added explicit escalation criteria for complex failures requiring manual review (test logic errors, security issues, performance regressions)
  • Updated documentation in docs/automation/overview.md with detailed autofix workflow improvements
  • Created comprehensive regression test suite (tests/regression/ci-autofix-improvements.test.ts) validating all enhancements
  • Addresses issue #132 - Review and improve copilot-ci-autofix workflow effectiveness

• 0.7.27 – 2025-10-26 (changelog)

  • Performance optimization documentation
  • Created comprehensive performance optimization guide in docs/operations/performance-optimization.md
  • Documented CPU optimization strategies including budget management, early termination, and caching patterns
  • Added memory management best practices covering cleanup, efficient data structures, and memory hygiene
  • Included pathfinding optimization techniques with reusePath values and cached pathfinding patterns
  • Documented profiling and monitoring approaches using existing PerformanceTracker, StatsCollector, and SystemEvaluator
  • Provided performance patterns and anti-patterns with code examples
  • Referenced integration with PTR monitoring infrastructure (#117, #299, #287)
  • Added links to existing regression tests for performance validation

• 0.7.25 – 2025-10-26 (changelog)

  • CPU optimization to maintain below 90% threshold
  • Reduced default CPU safety margin in BehaviorController from 90% to 80% for earlier creep processing cutoff
  • Reduced per-creep CPU threshold from 2.0 to 1.5 CPU to detect expensive operations earlier
  • Lowered PerformanceTracker warning threshold from 80% to 70% CPU usage
  • Lowered PerformanceTracker critical threshold from 95% to 90% CPU usage
  • Reduced Kernel emergency CPU threshold from 95% to 90%
  • Increased movement reusePath values from 5-20 ticks to 30-50 ticks to reduce pathfinding overhead
  • Added regression test suite to validate CPU optimization thresholds and prevent future performance degradation
  • These changes significantly reduce CPU consumption by minimizing expensive pathfinding operations

• 0.7.19 – 2025-10-25 (changelog)

  • Deploy workflow trigger mechanism (run #18800751206)
  • Updated deploy workflow to use workflow_run events instead of release events
  • Fixed version resolution logic to use git describe --tags --abbrev=0 for workflow_run triggers
  • Improved conditional logic to handle both workflow_run and workflow_dispatch events properly
  • Resolves regression tests expecting modernized CI/CD integration with Post Merge Release workflow

• 0.7.13 – 2025-10-25 (changelog)

  • Regression test failure for copilot-exec force-response parameter (run #18795077062)
  • Updated test to use correct workflow filenames after monitoring workflow consolidation
  • Changed screeps-stats-monitor.yml to screeps-monitoring.yml in test expectations
  • Removed reference to deleted copilot-autonomous-monitor.yml workflow
  • Test now correctly validates backward compatibility for force-response parameter

• 0.7.11 – 2025-10-24 (changelog)

  • Consolidated monitoring workflows for improved efficiency
  • Merged copilot-autonomous-monitor.yml and screeps-stats-monitor.yml into single screeps-monitoring.yml workflow
  • Combines autonomous strategic monitoring with PTR telemetry analysis in unified execution
  • Reduces workflow overhead from two parallel runs to one consolidated run every 30 minutes
  • Maintains all existing functionality: MCP server integration, PTR anomaly detection, strategic analysis, push notifications
  • Updated all documentation references to reflect consolidation
  • Updated copilot-ci-autofix.yml workflow trigger list
  • Created unified prompt template .github/copilot/prompts/screeps-monitor with 7-phase analysis pipeline
  • Post Merge Release workflow permission error (run #18794330724)
  • Excluded workflow files from prettier formatting in .prettierignore
  • Resolves GitHub rejection when pushing commits after workflows:write permission was removed
  • Prevents workflow file modifications during automated release process
  • Maintains security by avoiding workflows:write permission requirement
  • TypeScript type safety violations in fetch-screeps-stats test (run #18793984308)
  • Removed unnecessary eslint-disable comments that weren’t effective
  • Added proper TypeScript types to vitest mocks using ReturnType<typeof vi.fn>
  • Replaced (global.fetch as any) patterns with properly typed mockFetch variable
  • Added type assertions for mock.calls access patterns with explicit types like [string, RequestInit]
  • Used optional chaining for safer header access in assertions
  • Resolves linting failures that were blocking post-merge release workflow

• 0.7.1 – 2025-10-24 (changelog)

  • Incremental changelog management for version releases
  • Added releaseVersion() function to scripts/lib/changelog.ts for moving unreleased changes to version sections
  • Created scripts/release-changelog.ts CLI script to update CHANGELOG.md during version releases
  • Updated .github/workflows/post-merge-release.yml to automatically move unreleased changes to new version sections
  • Each version now contains only changes since the previous version (follows Keep a Changelog principles)
  • [Unreleased] section is automatically cleared after each version release
  • Added comprehensive unit tests (5 test cases) for changelog release functionality
  • Addresses issue: chore: implement incremental changelog management for version releases
  • TypeScript lint compliance: Removed unsafe any usage in automation scripts and tests
  • Added a typed Screeps raw API wrapper and stricter spawn placement flow in scripts/screeps-autospawn.ts
  • Declared a typed global Hexo reference for plugin loading and tightened test doubles to avoid unbound methods
  • Hardened mockup integration tests and helpers to dynamically import screeps-server-mockup without any casts
  • Hexo Documentation Build: Fixed markdown renderer loading in Hexo build script
  • Added proper plugin loading mechanism using global hexo variable for hexo-renderer-marked
  • Ensures markdown files are rendered to HTML instead of staying as .md files
  • Fixes documentation site deployment generating raw markdown files instead of HTML
  • Fixes run ID: 18781158449
  • Screeps Spawn Monitor Shard Parsing: Fixed shard/room parsing in autospawn script API calls
  • Parse shard name and room name from format “shard3/E45S25” for terrain and spawn placement API calls
  • Pass shard parameter correctly to roomTerrain() and placeSpawn() API methods
  • Fixes “Failed to get room terrain” error when spawn placement tries to analyze multi-shard rooms
  • Fixes run ID: 18780039750
  • Screeps Spawn Monitor API Failure: Fixed terrain API method call in autospawn script
  • Use correct api.raw.game.roomTerrain(roomName, 1) instead of invalid api.raw.game["room-terrain"]({ room, shard })
  • Removed unused shard parsing logic that was unnecessary for working API call
  • Fixes “api.raw.game.room-terrain is not a function” error
  • Fixes run ID: 18779690172
  • Screeps Spawn Monitor API Failure: Fixed incorrect API call for room terrain on sharded servers
  • Fixed screeps-api call to use correct room-terrain endpoint with proper shard parameter
  • Parse shard from room name format (shard3/E45S25) instead of passing invalid parameters
  • Resolves “Failed to get room terrain” error in autospawn workflow
  • Fixes run ID: 18779519651
  • Documentation Site Build Failure: Fixed duplicate dependencies in package.json causing lockfile conflicts in CI
  • Removed duplicate tsx and marked entries from dependencies section (keeping them in devDependencies)
  • Updated bun.lock to reflect proper dependency resolution
  • Fixes run ID: 18777257201
  • Daily Autonomous Bot Monitoring workflow
  • Created .github/workflows/copilot-autonomous-monitor.yml scheduled daily at 06:00 UTC
  • Comprehensive strategic analysis combining bot performance and repository health
  • Six-phase analysis pipeline: authentication, bot performance, repository health, strategic decisions, issue management, strategic reporting
  • MCP server integration via .github/mcp/screeps-mcp.json for bot console access using @ralphschuler/screeps-api-mcp
  • Direct Screeps console interaction for analyzing spawning, CPU, energy, RCL, defense, and strategic execution
  • GitHub repository analysis for codebase quality, automation effectiveness, and development velocity
  • Intelligent autonomous issue creation/update/close with evidence-based recommendations (up to 10 issues per run)
  • Bot health scoring (0-100) with top priorities and strategic recommendations
  • Safety controls: read-only by default, rate limiting (daily schedule, max 5 console commands per phase), prohibited destructive actions
  • Comprehensive documentation in docs/automation/autonomous-monitoring.md with architecture, usage, troubleshooting, and best practices
  • Updated README.md to include Autonomous Monitor Agent in agent types list
  • Addresses issue ralphschuler/.screeps-gpt#239 (autonomous monitoring and strategic automation)
  • Documentation restructuring for improved navigation
  • Created comprehensive docs/getting-started.md with detailed setup instructions, prerequisites, development commands, runtime architecture, and contributing workflow
  • Refactored README.md to focus on concise project overview emphasizing Copilot agent swarm concept
  • Added clear description of Screeps GPT as autonomous AI playground where multiple GitHub Copilot agents collaboratively develop a Screeps bot
  • Documented system architecture with three integrated layers: Runtime AI, Development Infrastructure, and AI Agent Orchestration
  • Included key features section highlighting autonomous agent swarm, CI/CD, self-evaluation, and documentation-first approach
  • Organized documentation links into categorized sections: Core, Technical, Monitoring & Operations, and Strategy & Development
  • Improved onboarding experience with clear navigation from README to detailed documentation
  • Addresses issue ralphschuler/.screeps-gpt#[issue_number] (documentation restructuring)
  • Docker containerization for development, testing, and building
  • Created Dockerfile.test with Node.js 20 + Python 2 for running test suites
  • Created Dockerfile.build with Node.js 20 for building the Screeps AI
  • Created Dockerfile.mockup with Node.js 16 + Python 2 for screeps-server-mockup compatibility
  • Added docker-compose.yml orchestrating dev, test, build, lint, and format services
  • Added .dockerignore to optimize Docker build context
  • Added Docker commands to package.json: docker:test:unit, docker:test:e2e, docker:test:mockup, docker:build:ai, docker:lint, docker:format, docker:dev, docker:shell
  • Created comprehensive Docker Development Guide at docs/operations/docker-guide.md
  • Updated README.md with Docker prerequisites, commands table, and contributing workflow
  • Provides isolated, reproducible development environments without local Node.js/Python installation
  • Enables simultaneous support for Node.js 20 (testing/building) and Node.js 16 (mockup tests)
  • Addresses issues #188 (Node.js migration), #204 (Bun integration), #200 (act CLI consistency)
  • Builder and remote miner creep roles
  • Registered new role definitions in the behavior controller with dedicated state machines and spawn configurations
  • Extended runtime types/memory helpers to track remote assignments deterministically
  • Added unit, e2e, and regression tests covering spawn logic plus travel/mine/return transitions
  • Documented strategy updates in docs/runtime/strategy/creep-roles.md and docs/runtime/strategy/scaling-strategies.md
  • Deterministic creep naming in BehaviorController
  • Replaced Math.random() with memory-persisted counter for creep name generation
  • Ensures deterministic AI behavior for reliable testing and debugging
  • Creep names now follow pattern: {role}-{game.time}-{counter} (e.g., harvester-100-0)
  • Documented the deterministic spawn naming scheme in docs/runtime/strategy/creep-roles.md so monitoring agents can trace counter resets
  • Added unit tests verifying deterministic naming behavior across test runs
  • Added regression test to prevent future Math.random() usage in runtime code
  • Resolves issue #174 and aligns with repository coding standards for deterministic runtime
  • Improves testing reliability and debugging consistency for autonomous AI validation
  • Documented Bun-first workflow and Node.js 18–22 support window
  • Updated README.md, AGENTS.md, DOCS.md, and docs/index.md to highlight Bun commands and supported Node versions.
  • Verified package.json engines and scripts align with the documented workflow.
  • Updated script messaging to reference bun run versions:update where applicable.
  • Updated package dependencies while maintaining Node.js 16 compatibility
  • Updated semver from 7.6.2 to 7.7.3 to address ReDoS security vulnerability (GHSA-c2qf-rxjj-qqgw)
  • Verified all build, lint, and test pipelines function correctly after update (66 tests passing)
  • Maintained Node.js 16.x compatibility as required by package.json engines field
  • Created comprehensive security assessment document at docs/security/dependency-vulnerabilities.md
  • Documented remaining 48 vulnerabilities: 79% are in optional testing dependencies, not production
  • Verified production bundle excludes all vulnerable dependencies (axios, lodash, angular, etc.)
  • Remaining vulnerabilities are acceptable risks per security assessment
  • Simplified Copilot model configuration
  • Removed .github/copilot/model-config.json file
  • Updated copilot-exec action to only pass --model flag when a model is explicitly specified
  • Model resolution now: input parameter → COPILOT_MODEL env var → Copilot CLI default
  • When no model is specified, Copilot CLI uses its own default model selection
  • Updated documentation in README.md and docs/automation/overview.md
  • Fixed Vitest CI failure in Node.js 16 environment (run 18742323437)
  • Improved crypto.getRandomValues() polyfill in tests/setup.ts to use randomBytes() instead of webcrypto
  • Fixes Vitest startup error: “TypeError: crypto.getRandomValues is not a function”
  • Resolves post-merge-release workflow failures where husky pre-commit hook failed during version bump
  • Maintains Node.js 16.x compatibility as required by package.json engines field
  • Node.js 16 doesn’t include Web Crypto API, but it’s required by Vite/Vitest
  • Uses Node.js built-in randomBytes() to implement the crypto polyfill
  • Ensures all test suites run successfully in CI workflows using Node.js 16
  • Node.js 16 compatibility for lint-staged in CI workflows
  • Downgraded lint-staged from v16.2.5 to v13.3.0 to maintain Node.js 16.14.0+ compatibility
  • Fixes post-merge-release.yml workflow failure caused by nano-spawn@2.0.0 dependency requiring Node.js 17+ (node:readline/promises)
  • Repository continues to use Node.js 16.x for native dependency compatibility (Python 2 requirement)
  • Removed unused @typescript-eslint/no-unsafe-return ESLint disable directive in tests/mockup/setup.ts
  • Verified no other dependencies have Node.js version incompatibilities
  • Push notification system for repository and Screeps bot monitoring
  • Integrated Push by Techulus API for real-time alerts on critical events
  • Created scripts/send-push-notification.ts with rate limiting and error handling
  • Added composite action .github/actions/send-push-notification for workflow integration
  • Implemented notifications in deploy workflow (success/failure alerts)
  • Implemented notifications in quality-gate workflow (build failure alerts)
  • Implemented PTR monitoring alerts via scripts/check-ptr-alerts.ts
  • Added automated notifications for high CPU usage (>80% sustained), critical CPU (>95%), and low energy
  • Created comprehensive documentation at docs/automation/push-notifications.md
  • Added unit tests for notification utility with 100% coverage
  • Rate limiting: 5 second minimum interval, max 10 notifications per minute
  • Graceful degradation: notification failures never break workflows
  • Secure implementation: PUSH_TOKEN stored as GitHub secret, no sensitive data in notifications
  • Complements existing email notification system (#134)
  • Integrates with PTR monitoring infrastructure (#152, #117)
  • ESLint flat config migration with Node 16 structuredClone polyfill (#156)
  • Migrated from deprecated .eslintrc.cjs to modern eslint.config.mjs flat configuration format
  • Added .eslintrc-polyfill.cjs to provide structuredClone implementation for Node.js 16.x compatibility
  • Updated npm lint scripts to use flat config (removed ESLINT_USE_FLAT_CONFIG=false)
  • Updated lint-staged configuration to use simplified ESLint commands
  • Preserved all existing ESLint rules and TypeScript plugin configurations
  • Resolves ESLint deprecation warnings for v10.0.0 preparation
  • Fixes ReferenceError: structuredClone is not defined when running ESLint with @typescript-eslint v8+ on Node 16
  • Configurable Copilot model selection (#146)
  • Removed hardcoded gpt-5 default from copilot-exec action
  • Created centralized model configuration file (.github/copilot/model-config.json) with default gpt-4.1
  • Implemented priority-based model resolution: input parameter → COPILOT_MODEL env var → config file → hardcoded default
  • Added model validation and logging to copilot-exec composite action
  • Updated documentation (README.md, docs/automation/overview.md) with configuration examples
  • Supports repository-level and workflow-level model overrides via environment variables
  • Maintains backward compatibility with explicit model parameters in workflows
  • Copilot exec pre-checkout + dependency caching optimisation (#101)
  • Added detection & conditional checkout to composite action (skips if .git present)
  • Implemented node_modules cache keyed by OS + package-lock.json hash with restore keys fallback
  • Added conditional npm ci install only on cache miss and when package.json exists
  • Added total duration metric step for performance tracking
  • Backward compatible: existing workflows need no changes
  • Standardized label system with state, type, and priority categories (#138)
  • Implemented three-tier labeling system for improved issue and PR management
  • Added state labels: state/pending, state/backlog, state/in-progress, state/blocked, state/canceled, state/done
  • Added type labels: type/bug, type/feature, type/enhancement, type/chore, type/question
  • Added priority labels: priority/critical, priority/high, priority/medium, priority/low, priority/none
  • Added workflow labels: good-first-issue, help-wanted, wontfix, duplicate, invalid
  • Preserved existing domain labels (automation, runtime, documentation, monitoring, dependencies, regression)
  • Preserved process labels (Todo, monitoring, copilot, needs/regression-test)
  • Updated all issue templates to use new label structure
  • Updated all copilot prompts to use new labels (issue-triage, stats-analysis, repository-audit, ci-autofix, email-triage)
  • Updated documentation (README.md, AGENTS.md, copilot-instructions.md) with comprehensive label guidance
  • Created comprehensive label system guide at docs/automation/label-system.md
  • Maintained backward compatibility by keeping deprecated labels (bug, enhancement, severity/*) marked as deprecated
  • Modernized CI/CD deployment workflow to use standard GitHub DevOps practices (#126)
  • Replaced manual release PR creation with automated semantic versioning based on conventional commits
  • Post-merge workflow now commits version bumps directly to main instead of creating release branches and PRs
  • Implemented semantic version bumping: feat: → minor, fix:/chore: → patch, BREAKING CHANGE: → major
  • GitHub Releases are now created automatically using GitHub’s native API with auto-generated release notes
  • Deploy workflow now uses GitHub’s production environment for deployment protection and tracking
  • Deploy workflow triggers on both version tags and GitHub Release published events
  • Removed workflow_run trigger complexity in favor of native release events
  • Added version:bump-semantic npm script and scripts/bump-version-semantic.ts for semantic versioning
  • Updated documentation in README.md and docs/automation/overview.md to reflect new CI/CD workflow
  • Fixed vitest dependency conflict in Deploy Screeps AI workflow by upgrading @vitest/coverage-v8 from ^0.33.0 to ^3.2.4 for compatibility with vitest ^3.2.4 (workflow run: 18705052117)
  • Fixed email triage workflow not creating issues by removing contradictory JSON formatting in prompt template (#115)
  • Fixed CI failure in npm run versions:update by adding missing trailing newline to docs/changelog/versions.md (regression test: tests/regression/versions-file-trailing-newline.test.ts, workflow run: 18703566323)
  • Fixed git push conflict in post-merge release workflow by adding remote ref updates before commit operations (regression test: tests/regression/post-merge-workflow-git-race-condition.test.ts, workflow run: 18703919715)
  • Fixed automatic Todo label assignment in issue triage by removing Todo from automatic labeling per issue #78 to prevent unwanted automation triggers
  • Specialized PR templates for Copilot automation workflows (#130)
  • Created .github/PULL_REQUEST_TEMPLATE/copilot-todo.md for Todo workflow PRs with automation-specific checklists
  • Created .github/PULL_REQUEST_TEMPLATE/copilot-quickfix.md for CI autofix PRs with validation-focused content
  • Templates reduce cognitive load by removing irrelevant manual checklist items
  • Improved reviewer guidance specific to Copilot-generated changes
  • Maintained quality standards while tailoring context to automated change types
  • Ready workflow integration (requires manual application due to workflow permission constraints)
  • Enhanced Copilot prompt templates with action enforcement rules (#127)
  • Added mandatory action requirements with explicit “MUST” criteria for all workflows
  • Implemented comprehensive failure handling for GitHub API issues, missing data, and timeout conditions
  • Added explicit output quality requirements and validation criteria
  • Included actionable finding criteria and severity assessment guidelines
  • Added pre/post-execution validation steps for all automated operations
  • Standardized prompt template naming and structure
  • Renamed todo-issue → todo-automation for consistency with workflow purpose
  • Renamed repository-audit → repository-review for clarity
  • Updated corresponding workflow files to reference new prompt paths
  • Enhanced action appropriateness criteria
  • Added explicit guidelines for when automatic fixes are appropriate vs. manual intervention required
  • Implemented quality gates preventing inappropriate automation of complex issues
  • Added concrete thresholds and examples for anomaly detection and severity assessment
  • Comprehensive prompt template audit documentation in docs/automation/prompt-audit.md
  • Detailed analysis of existing templates with strengths and gaps identified
  • Enhancement framework and recommendations for consistent action enforcement
  • Impact assessment and validation requirements for template changes
  • Created .github/copilot-instructions.md with repository-specific guidelines for GitHub Copilot coding agent
  • Includes coding standards, development workflow, testing expectations, and documentation requirements
  • References comprehensive documentation in AGENTS.md, README.md, and docs/ for detailed guidance
  • Integrated Screeps API MCP server with GitHub Copilot workflows for direct Screeps server interaction
  • Integrated Playwright MCP server for browser automation capabilities
  • Added MCP server configuration files: .github/mcp/screeps-api.json and .github/mcp/playwright.json
  • Enhanced copilot-exec action to support multiple MCP servers via additional-mcp-config parameter
  • Added comprehensive MCP server documentation in AGENTS.md and docs/automation/overview.md
  • Implemented a basic Screeps runtime with headcount-based spawning and simple harvester/upgrader state machines
  • Migrated from pnpm to npm as the package manager
  • Updated all workflow files to use npm instead of pnpm
  • Updated documentation to reference npm commands
  • Added .nvmrc file to specify Node.js 16.20.2
  • Updated package.json to remove pnpm references and specify npm in engines
  • Copilot CI AutoFix workflow now monitors all workflow failures (except itself) instead of only Quality Gate failures, enabling automated fixes for any CI failure
  • Updated screeps-stats-monitor.yml to use Screeps API MCP server for direct telemetry access
  • Enhanced .github/copilot/prompts/stats-analysis to document available MCP servers
  • Updated README.md secrets documentation to include MCP authentication variables
  • Enhanced Copilot Todo automation workflow to create draft pull requests immediately and show visible implementation progress
  • Draft PRs are created at the start of the automation process for transparency
  • Implementation progress is shown through frequent commits and PR description updates using the report_progress tool
  • Users can follow along with the implementation in real-time
  • PRs are marked as ready for review only after all validation passes
  • Updated .github/copilot/prompts/todo-issue with new draft PR workflow
  • Updated documentation in README.md, docs/automation/overview.md, and AGENTS.md
  • Fixed build error caused by node-gyp attempting to use Python 2 syntax with Python 3
  • Moved @screeps/common, @screeps/driver, @screeps/engine, and screeps-server-mockup packages to optionalDependencies to allow installation to succeed even when native modules fail to build
  • Added .npmrc to configure build behavior for optional dependencies
  • Mockup tests now gracefully skip when isolated-vm build fails (as documented in tests/mockup/README.md)
  • Fixed post-merge workflow recursive execution by adding condition to skip when release PRs are merged back to main
  • Fixed deployment workflow not triggering by adding workflow_run trigger to chain deployment after successful release preparation
  • Simplified tag creation in post-merge workflow from GitHub API to git commands for clarity
  • Deploy workflow now falls back to latest version tag when no tag exists on triggering commit (resolves workflow run 18701965424)
  • Fixed deployment failure with empty environment variables - Changed deploy script to use || instead of ?? operator so empty string secrets default to proper Screeps API values (resolves workflow run 18702433741)

• 0.1.0 – 2024-06-01 (changelog)

  • Added a curated docs/ knowledge base and updated automation guidance so every fix documents findings and regression coverage.
  • Extended the shared copilot-exec action to support GitHub MCP configuration and fuel new prompt templates for stats analysis and CI auto-fixes.
  • Introduced scheduled Screeps stats monitoring, label synchronisation, and Copilot-driven CI auto-fix workflows with supporting scripts and prompts.