Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
94 lines (68 loc) · 2.3 KB

CONTRIBUTING.md

File metadata and controls

94 lines (68 loc) · 2.3 KB

Contributing to AKF

Thank you for your interest in contributing to the Agent Knowledge Format project!

Prerequisites

  • Python 3.9+
  • Node.js 18+
  • git

Development Setup

git clone https://github.com/HMAKT99/AKF.git
cd AKF

# Python SDK
cd python && pip install -e ".[dev]" && cd ..

# TypeScript SDK
cd typescript && npm install && cd ..

# Website
cd site && npm install && cd ..

Project Structure

Directory Description
python/akf/ Python SDK
typescript/src/ TypeScript SDK
site/ Website (Vite + React + Tailwind)
spec/ Format specification and JSON schema
packages/ Framework integrations (MCP, LangChain, LlamaIndex, CrewAI)
extensions/ VS Code, GitHub Action, Office, Google Workspace
skills/ Agent skill files

Running Tests

# Python
cd python && python -m pytest tests/ -v

# TypeScript
cd typescript && npm test

# Website (build check)
cd site && npm run build

PR Process

  1. Create a feature branch (git checkout -b feature/my-feature)
  2. Make your changes
  3. Run all relevant tests (see above)
  4. Add tests for new functionality
  5. Open a PR against main

Reporting Bugs

Open a GitHub issue with:

  • AKF version (akf --version)
  • Python/Node.js version
  • Minimal reproduction case
  • Expected vs actual behavior

Feature Requests

Open a GitHub issue with the "feature request" label. Include:

  • Use case description
  • Proposed API surface
  • Example code showing how you'd use it

Code Style

  • Follow existing patterns in the codebase. No unnecessary abstractions.
  • Python: Use type hints. Run pytest before submitting.
  • TypeScript: Use strict TypeScript. Run npm test before submitting.
  • Spec changes: Must maintain backward compatibility with existing .akf files.

Areas for Contribution

  • Framework integrations (LangChain, LlamaIndex, CrewAI, etc.)
  • New file format handlers in universal.py (video, audio, and beyond)
  • Additional compliance regulation checks
  • Documentation and examples
  • IDE extensions
  • Multi-agent features: delegation policies, team streaming, A2A protocol bridge
  • Agent identity: AgentCard registry, cross-platform discovery

License

By contributing, you agree that your contributions will be licensed under the project's MIT License.