Skip to main content

CODITECT Core - Master Project Plan

Product: AZ1.AI CODITECT - Distributed Intelligence Framework Repository: coditect-core (Primary Product / CODITECT Brain) Owner: AZ1.AI INC. Version: 1.7.2 (Core) + 2.0 (UAF in development) Status: Production-ready → Evolution to Full Autonomy Last Updated: December 22, 2025

Quick Navigation​

For Active Roadmap: See v2/ folder for epic-based project structure For Current Status: See status/PROJECT-STATUS.md For Task Tracking: See tasklists/ACTIVE-TASKLIST.md

Executive Summary​

CODITECT Core is the foundational distributed intelligence framework that serves as the "brain" for the AZ1.AI CODITECT platform. This repository contains the complete agent orchestration system, command framework, skills library, and training materials that enable autonomous AI-powered project development from concept to production.

Product Classification​

This is AZ1.AI's PRIMARY PRODUCT - the CODITECT brain that powers:

  • Local-first installation (no cloud dependency)
  • Optional cloud sync and collaboration
  • Distributed intelligence architecture via .coditect symlink chain
  • Complete project lifecycle automation
  • Zero catastrophic forgetting via context database system
  • Comprehensive operator training (4-6 hour certification)

Current Status (December 2025)​

Framework Maturity: See config/component-counts.json for current component totals

  • Core Framework: Production-ready (v1.7.2)
  • Universal Agent Framework: Active development (v2.0)
  • Agent System: 122 specialized agents across 8 domains
  • Command System: 134 slash commands operational
  • Skills Library: 179 production skills catalogued
  • Scripts: 206+ Python/shell automation scripts
  • Hooks: 38 event-driven automation triggers
  • Workflows: 1,149 documented workflows
  • Training System: Complete 4-6 hour certification path
  • Installation: Cross-platform automated installers

Recent Milestones:

  • Dec 21: Documentation reorganization (customer/contributor split)
  • Dec 20: A2A Protocol integration (122 agent cards, 67 new skills)
  • Dec 17-19: Pilot launch infrastructure (license server, OAuth, Cloud Run)
  • Dec 11: Onboarding system + test suite (78 tests)
  • Nov 29: Multi-session integration + Docker development environment

Architecture Overview​

Distributed Intelligence Architecture​

Core Concept: .coditect symlink chain enables intelligence at every project node

Master Repository (coditect-rollout-master)
├── .coditect/ (git submodule → coditect-core)
│   ├── agents/          # 122 specialized AI agents
│   ├── commands/        # 134 slash commands
│   ├── skills/          # 179 production skills
│   ├── scripts/         # 206+ automation scripts
│   ├── docs/            # Customer documentation
│   ├── internal/        # Contributor documentation
│   └── context.db       # Anti-forgetting memory system
│
├── .claude -> .coditect (symlink for Claude Code compatibility)
│
└── submodules/ (57 submodules with distributed intelligence)

Why This Matters:

  • Every submodule has access to full CODITECT intelligence
  • Run Claude Code from any directory → access all agents, commands, skills
  • Single source of truth (git submodule updates propagate automatically)
  • Resilience: damage to one node doesn't disable system
  • Scalability: add nodes without redesigning architecture

Strategic Positioning​

Commercial Product Strategy​

  1. Local-First Open Core - Free local installation, commercial cloud features
  2. Training & Certification - Revenue from operator training programs
  3. Enterprise Features - Team collaboration, advanced analytics, priority support
  4. Platform Services - Cloud sync, marketplace, analytics dashboard

Market Differentiation​

  • Only distributed intelligence framework for project development
  • Zero vendor lock-in - works locally without cloud dependency
  • Context database system eliminates catastrophic forgetting
  • Complete training system - 4-6 hour certification vs. competitors' weeks
  • Multi-LLM compatibility - not locked to single AI provider

Historical Context: Phase-Based Development (2024-2025)​

Phase 1: Foundation (2024 Q4) ✅ COMPLETE

  • Initial agent framework (50 agents)
  • Basic command system (40 commands)
  • Core documentation structure

Phase 2: Documentation & Skills (2025 Q1) ✅ COMPLETE

  • Comprehensive documentation (60+ guides)
  • Skills library standardization (100+ skills)
  • Training materials creation

Phase 3: Automation & Integration (2025 Q2) ✅ COMPLETE

  • Git workflow automation
  • Session preservation system
  • Multi-session integration

Phase 4: Production Readiness (2025 Q3) ✅ COMPLETE

  • Docker development environment
  • CI/CD deployment workflows
  • Component activation system

Phase 5: Universal Agent Framework (2025 Q4) 🚧 IN PROGRESS

  • A2A Protocol integration
  • Enhanced agent interoperability
  • LLM Council consensus patterns
  • Pilot launch infrastructure

Current Approach: Epic-based development (see v2/ folder)

  • More flexible than phase-based planning
  • Better alignment with Agile/Sprint workflows
  • Easier to track progress and dependencies

Component Activation from v2/E001​

Epic E001: Autonomous Agent Orchestration contains the complete roadmap for achieving full framework autonomy.

Critical Gap for Full Autonomy:

  • ❌ Inter-agent communication (agents cannot send tasks to each other)
  • ❌ Message Bus infrastructure (RabbitMQ/Redis)
  • ❌ Agent Discovery Service (capability-based routing)
  • ❌ Task Queue Manager (persistent queue with dependencies)
  • ❌ Circuit Breaker and resilience patterns
  • ❌ Comprehensive test coverage (currently <50%)
  • ❌ Production monitoring and observability

See: v2/epics/E001-AUTONOMY/ for detailed implementation plan

Success Metrics (Target State)​

MetricCurrentTargetStatus
AutonomyManual orchestration95% autonomousSee v2/E001
Framework Maturity78%100%Q2 2026
Test Coverage45%80%Q1 2026
Component Count1,8332,000+Q2 2026
Active Users500+2,000+Q2 2026
GitHub Stars2501,000+Q2 2026

Quality Gates & Testing​

Current Testing Strategy:

  • Unit tests: 78 tests (scripts/test-suite.py)
  • Integration tests: Manual validation workflows
  • Documentation tests: Cross-reference validation
  • Component activation: Manual verification

Target Testing Strategy (v2/E002):

  • Unit test coverage: 80%+
  • Integration tests: Automated CI/CD pipeline
  • End-to-end tests: User workflow automation
  • Performance tests: Load testing and benchmarking

Budget & Resource Requirements​

Current Monthly Operating Cost: ~$300

  • Infrastructure: $100/month (Cloud SQL, Cloud Run)
  • Tools & Services: $150/month (GitHub, monitoring)
  • Support: $50/month (community management)

Pilot Launch Infrastructure:

  • Cloud Run: $33/month (initial pilot)
  • Cloud SQL: $50/month (db-f1-micro)
  • Stripe: 2.9% + $0.30 per transaction
  • Total estimated: $100-150/month for 50-100 pilot users

Production Scale (10K+ users):

  • GKE Autopilot: $505/month (3-node cluster)
  • Cloud SQL: $200/month (db-n1-standard-2)
  • Cloud Run: Minimal (auth/webhooks only)
  • Total estimated: $1,000-1,500/month

Risks & Mitigation​

Technical Risks​

RiskProbabilityImpactMitigation
Component activation complexityHighMediumImproved UX, better documentation
Session preservation failuresLowHighMultiple backup methods, robust error handling
Scalability issuesLowMediumStreaming architecture, load testing
Multi-LLM compatibilityMediumHighRegular testing across providers

Organizational Risks​

RiskProbabilityImpactMitigation
Limited resourcesMediumMediumEfficient automation, community contributions
Documentation debtLowHighAutomated generation, regular reviews
User adoption barriersMediumMediumEnhanced training, better onboarding
CompetitionLowLowUnique architecture, strong documentation

Long-Term Evolution​

Q1 2026 (January - March)​

  • Component activation UI/dashboard
  • Enhanced session analysis workflows
  • Production deployment templates
  • UAF v2.0 beta preparation

Q2 2026 (April - June)​

  • UAF v2.0 beta release
  • Multi-user collaboration features
  • Distributed execution engine
  • Enterprise features (RBAC, audit logging)

Q3-Q4 2026​

  • UAF v2.0 production release
  • Advanced AI capabilities
  • Platform integrations (VS Code, JetBrains, etc.)
  • Community marketplace

Dependencies & Integration​

Critical Dependencies:

  • Python 3.10+ (core runtime)
  • Git 2.25+ (submodule support)
  • Claude Code / LLM providers (AI execution)
  • Docker (optional development environment)

Platform Integrations:

  • GitHub (version control, CI/CD)
  • Google Cloud Platform (production infrastructure)
  • Stripe (billing and subscriptions)
  • Multiple LLM providers (Claude, GPT, Gemini, etc.)

Additional Resources​

Planning Documents​

Implementation Guides​

  • v2/epics/ - 10 epic folders with tasklists
  • v2/sprints/ - Sprint planning documents
  • Component activation: See v2/epics/E001-AUTONOMY/

Documentation​

  • Customer docs: /docs/ (getting started, guides, reference)
  • Contributor docs: /internal/ (architecture, research, testing)

Document Status: Active (Consolidated) Last Updated: December 22, 2025 Next Update: As needed (v2/ is primary planning source) Owner: Hal Casteel, Founder/CEO/CTO, AZ1.AI INC