MCAF is a framework for building software products together with AI coding agents.
It defines how to:
keep all relevant context in a single repository
describe work clearly before coding
use integration/API/UI tests, containerized environments, and static analysis as hard gates
encode repository-specific rules for AI agents in AGENTS.md at the repo root
The goal of MCAF:
Use AI to build real products in a way that is predictable, safe, and repeatable.
MCAF has three core elements:
Context — code, tests, documentation, and AGENTS.md live together in the repository.
Verification — behaviour and code quality are proven by automated tests and static analysis, not opinions.
Instructions — AGENTS.md defines how AI agents work in this repo and how they learn from feedback.
This Guide defines the framework.
The root AGENTS.md in each repository applies it to that codebase.
MCAF is a framework for building software products together with AI coding agents.
It defines how to:
keep all relevant context in a single repository describe work clearly before coding use integration/API/UI tests, containerized environments, and static analysis as hard gates encode repository-specific rules for AI agents in AGENTS.md at the repo root The goal of MCAF:
Use AI to build real products in a way that is predictable, safe, and repeatable. MCAF has three core elements:
Context — code, tests, documentation, and AGENTS.md live together in the repository. Verification — behaviour and code quality are proven by automated tests and static analysis, not opinions. Instructions — AGENTS.md defines how AI agents work in this repo and how they learn from feedback. This Guide defines the framework. The root AGENTS.md in each repository applies it to that codebase.