managedcodemanagedcodeOfficial·5 Agent Skills Included

MCAF

Structured engineering governance for predictable AI-assisted software delivery

Standardizes how coding agents plan, build, test, and review software inside a repository. Covers feature specs, architecture decisions, testing, code review, CI/CD, security, and observability with traceable requirements. Removes guesswork, undocumented decisions, and unverified code by enforcing clear rules and verification gates. Helps teams ship reliable software faster with repeatable, auditable workflows.
npx skills add managedcode/MCAF --all -g -y
Available:

Gives your AI agent a project-wide operating manual covering architecture rules, verification commands, skill routing, and how to plan and delegate coding work safely.

All Skills in This Repository (5)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install MCAF?

Run `npx skills add managedcode/MCAF --all -g -y` in your terminal to install all skills in this suite globally.

What problem does MCAF solve?

It stops AI coding agents from producing unplanned, untested, or undocumented changes by enforcing feature specs, architecture decisions, and verification gates in the repository.

Does MCAF work with Claude Code and Codex?

Yes. All skills follow the universal SKILL.md standard and run in Claude Code, Codex, Cursor, and other compatible agent runtimes.

Can MCAF be used in a .NET project?

Yes. MCAF includes dedicated guidance for .NET solutions, including test framework selection, quality gates, and links to the external Managed Code .NET skill catalog.

Do I need to change my repository structure to use MCAF?

MCAF recommends a single repository with feature-based vertical slices, but its templates let you document existing layouts and plan gradual migrations.

Related Repositories in Software Engineering

View All in Software Engineering