xfarch/aikit

Claude plugin for working with the XF Architecture Model (CFAM) from AI assistants. Bundles skills, instructions and helpers so Claude can classify components, scaffold artefacts, review compliance, and answer questions about the XF spec.

Status

This repository is in early scaffolding. The initial commit contains only governance and tooling files (license, code of conduct, security policy, CI, contribution guidelines). The plugin manifest and skills will land in follow-up commits.

What's in scope

• A Claude plugin (plugin.json) declaring skills, prompts and MCP wiring.
• A library of skills that teach Claude how to apply XF rules in conversation: xf-classify, xf-explain, xf-review, xf-scaffold.
• Reusable instructions (AGENTS.md, CLAUDE.md, INSTRUCTIONS.md) consumable by any agent that wants XF-aware behaviour.

What's NOT in scope

• The normative XF specification — see xfarch/docs.
• TypeScript reference implementation — see xfarch/lib-npm.
• CLI validator / linter — see xfarch/tools.

License

MIT — see LICENSE.

Security

Please report vulnerabilities privately. See SECURITY.md.

Code of conduct

This project follows the Contributor Covenant v2.1.