GAAI-framework
Turns AI coding tools into reliable software delivery systems. Drop a .gaai/ folder into any project — Discovery defines what to build, Delivery executes autono
What is GAAI-framework?
Turns AI coding tools into reliable software delivery systems. Drop a .gaai/ folder into any project — Discovery defines what to build, Delivery executes autonomously until criteria pass. Works with Claude Code, Codex CLI, Gemini CLI, Cursor, and more. No SDK. No package. Markdown + YAML + bash.
Overview of GAAI-framework
What it is
GAAI-framework is a governance layer for AI coding tools. It is a .gaai/ folder that you drop into any git project. It uses Markdown, YAML, and bash, with a vendored Python runtime and a few Node helpers. There is no SDK, no package to install, and no external services. The framework turns AI coding tools into agentic software delivery systems by adding cross-session memory, scope authorization, and structured delivery.
Key features
- Discovery and Delivery isolation: Discovery reasons but never executes; Delivery executes but never decides scope. They run in separate OS processes so system prompts cannot contaminate each other.
- Backlog as contract: The backlog is the contract between Discovery and Delivery. Nothing gets built that is not in it.
- Delivery Daemon:
/gaai-daemonlaunches a background daemon that polls the backlog and delivers multiple Stories in parallel, each in its own tmux session. Requires a git repo with astagingbranch. - Portable to sub-agent-capable runtimes: The delivery workflow works with Claude Code, Codex CLI, Gemini CLI, Cursor, and more. Claude Code is the reference implementation.
- No SDK or package: Governance is Markdown, YAML, and bash. Delivery also needs
python3andnodeon PATH, plus the vendored PyYAML runtime shipped in.gaai/core/vendor/. - Structured project folder: The
.gaai/folder containscore/(framework engine, auto-synced) andproject/(your project data, never overwritten by updates).
Use cases
GAAI is built for developers who already have product clarity — solo founders, senior engineers, and small teams who know what to build and need an agent that ships it reliably without going off-script. It is for anyone who has said “the agent broke something it wasn’t supposed to touch.”
Getting started
Copy the .gaai/ folder into your project. Download from GitHub, drop .gaai/ into your project root, and tell your AI tool: “Read .gaai/core/README.md and bootstrap this project.”Option A — Ask your AI tool to do it:Install the GAAI framework into my current project.Determine {user-tool} by identifying which AI coding tool is running thisprompt. Valid values: claude-code | cursor | windsurf | other.If you cannot determine it, ask the user before proceeding.Then run: rm -rf /tmp/gaai git clone https://github.com/Fr-e-d/GAAI-framework.git /tmp/gaai bash /tmp/gaai/.gaai/core/scripts/install.sh --target . --tool {user-tool} --yes rm -rf /tmp/gaaiAfter install, show the user the next steps exactly as printed by theinstaller.Option B — CLI:bashgit clone https://github.com/Fr-e-d/GAAI-framework.git /tmp/gaai && bash /tmp/gaai/.gaai/core/scripts/install.sh --wizard && rm -rf /tmp/gaai
License and pricing
The license is not stated. The project is listed as open source.
Based on the project's public README and repository metadata.