Stop fighting your AI assistant. Start governing it.
Tired of AI-generated code that's buggy, inconsistent, and ignores your architecture? The AI Governor Framework is a safe, plug-and-play system designed to teach your AI your project's unique DNA. It provides a set of rules and workflows to turn any AI assistant from a chaotic tool into a disciplined engineering partner that respects your architectural decisions, best practices, and non-negotiable constraints.
Reclaim control. Enforce your coding standards. Ship with confidence.
This approach is rooted in one core principle: Context Engineering.
This isn't about bigger prompts or dumping your entire codebase into one, which is both ineffective and expensive. It's about giving the AI the right information at the right time. This framework achieves that by building a knowledge base of robust rules (the orders) and architectural READMEs (the context) that the AI consults on-demand.
This framework is built on a simple, robust principle: Treat your project's knowledge base like your codebase.
We leverage an In-Repo approach, keeping all governance rules and architectural context directly inside the repository. This makes the AI's knowledge base:
- Simple & Efficient: Zero network latency and no complex external systems.
- Evolutive & Maintainable: The AI's context evolves in lock-step with your code.
- Auditable & Versioned: Every change is tracked in
git, providing a clear, human-readable history.- Portable & Robust: Any developer can clone the repo and have the full, up-to-date context instantly, ensuring stability and consistency.
For complex external documentation, such as third-party APIs or external library, this in-repo system can be seamlessly combined with a RAG-based MCP server, such as Context7, to fetch and inject that external knowledge on demand. This leverages the best of both worlds: robust and versioned in-Repo governance for your internal architecture, and dynamic, on-demand context for external dependencies.
This is how we shift from the endless loop of prompting and fixing to strategic Governing.
The AI Governor Framework is composed of two distinct but complementary parts:
| Component | What It Is | How It's Used |
|---|---|---|
The Governance Engine (/rules) |
A set of powerful, passive rules that run in the background. | Your AI assistant discovers and applies these rules automatically to ensure quality and consistency. |
The Operator's Playbook (/dev-workflow) |
A set of active, step-by-step protocols for the entire development lifecycle. | You manually invoke these protocols to guide the AI through complex tasks like planning and implementation. |
The dev-workflow is now reinforced with a comprehensive suite of review protocols for code quality, security, and architecture. This new system is inspired by the successful, battle-tested workflows used by Anthropic's own team to build products like Claude Code.
Key changes include:
- A new, 7-layer quality audit protocol for robust validation.
- Simplified and clarified all workflow steps for better readability.
- Rewrote all documentation to focus on universal software engineering principles.
By adopting these proven strategies for structured AI collaboration, the framework provides a more robust and predictable path from idea to production-ready code.
The framework is built around a series of sequential protocols that move a feature from idea to production with clarity and control:
- 0️⃣ Bootstrap: Turns a generic AI into a project-specific expert. (One-Time Setup)
- 1️⃣ Define: Transforms an idea into a detailed PRD.
- 2️⃣ Plan: Converts the PRD into a step-by-step technical plan.
- 3️⃣ Implement & Review: Executes the plan, followed by a mandatory quality audit (
4-quality-audit.md) and a process retrospective (5-implementation-retrospective.md) to continuously improve the system.
Ready to install the framework and run your first governed task?
This guide provides a safe, non-destructive process to integrate the framework into any project.
1. Clone the Framework
Open a terminal at the root of your project and run the following command:
git clone https://github.com/Fr-e-d/AI-Governor-Framework.git .ai-governorThis downloads the entire framework into a hidden .ai-governor directory within your project.
2. Configure for Your Assistant
The final step depends on your AI assistant. Choose the relevant section below.
To get the best experience, including custom commands like /review, copy both the rules and the prompts into your project's .cursor directory.
# Copy governance rules
mkdir -p .cursor/rules && cp -r .ai-governor/rules/* .cursor/rules/
# Copy custom prompts for an enhanced workflow
mkdir -p .cursor/prompts && cp -r .ai-governor/.cursor/prompts/* .cursor/prompts/To enable custom commands like /review, copy the command definitions into your project's .claude directory.
mkdir -p .claude/commands && cp -r .ai-governor/.claude/commands/* .claude/commands/The framework is ready to use without extra steps. Your assistant will find its boot sequence (CLAUDE.md or OpenCode.md) and the necessary rules within the .ai-governor directory.
If you find this framework valuable, please consider showing your support. It is greatly appreciated!
This framework is an enhanced and structured adaptation inspired by the foundational work on AI-driven development by snarktank/ai-dev-tasks.
It is shared under the Apache 2.0 License. See the LICENSE file for more details. For contribution guidelines, please see CONTRIBUTING.md.
