Summary

ArcKit is an Enterprise Architecture Governance and Vendor Procurement Toolkit designed to transform scattered documentation into a systematic, AI-assisted workflow for enterprise architects. It addresses the challenges of inconsistent governance, manual vendor evaluation, and lost traceability by providing structured processes and AI-powered research and document generation. The toolkit benefits users by streamlining complex architecture lifecycle management, ensuring compliance, and enabling data-driven decision-making.

How It Works

ArcKit employs a command-driven, template-based approach augmented by AI agents and integrated knowledge bases (MCP servers). It automates governance through hooks for session initialization, project context injection, filename enforcement, output validation, and impact scanning. Autonomous research agents perform tasks like technology research, cloud service evaluation, and data source discovery. The system guides users through a structured workflow, from establishing principles and analyzing stakeholders to conducting design reviews and managing vendor procurement, ensuring comprehensive artifact generation and traceability.

Quick Start & Requirements

Installation varies by AI assistant:

• Claude Code: Install via plugin marketplace (/plugin marketplace add tractorjuice/arc-kit). Premier experience with all features.
• Gemini CLI: Install extension (gemini extensions install https://github.com/tractorjuice/arckit-gemini).
• GitHub Copilot (VS Code): Install CLI (pip install git+https://github.com/tractorjuice/arc-kit.git), then scaffold prompts (arckit init my-project --ai copilot).
• Codex/OpenCode CLI: Install CLI (pip install git+https://github.com/tractorjuice/arc-kit.git), then initialize (arckit init my-project).

Requires Python 3.11+ and a compatible AI coding agent (Claude Code v2.1.112+, Gemini CLI, OpenCode CLI, or OpenAI Codex CLI). Setup is generally zero-config after integrating with the chosen AI assistant.

Highlighted Details

• UK Government Compliance: Dedicated commands for Technology Code of Practice (TCoP), GDS Service Standard assessment, Secure by Design, AI Playbook, and MOD-specific security requirements.
• Strategic Analysis: Integrated Wardley Mapping for component evolution analysis, build vs. buy decisions, and strategic positioning.
• Visual Documentation: Generates architecture diagrams (C4 Model, Deployment, Sequence, Data Flow) using Mermaid syntax.
• ServiceNow Integration: Bridges architecture artifacts to operations with CMDB design, SLA definitions, and ITIL-aligned runbooks.
• AI-Assisted Workflows: Leverages AI for document generation, research, and analysis across the EA lifecycle.

Maintenance & Community

ArcKit is actively maintained, with the latest release noted as v4.6.6. While specific community links (Discord/Slack) or prominent contributor details are not explicitly highlighted in the README, the project structure and extensive command set suggest ongoing development.

Licensing & Compatibility

The project is released under the MIT License, permitting commercial use and integration into closed-source projects.

Limitations & Caveats

Native Windows support has limitations for certain hooks, recommending WSL2 or devcontainers for optimal experience. Large document generation may encounter AI token limits, with workarounds provided for specific commands and AI plans. Some experimental features, like Wardley Mapping sub-commands and certain cloud research commands, are in early development stages.