Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
34 lines (28 loc) · 1.49 KB

CONTRIBUTING.md

File metadata and controls

34 lines (28 loc) · 1.49 KB

Contributing to 100 AI-Native Python Projects

Thank you for your interest in building the pattern library for the human-in-the-loop era!

Philosophy

This is not a script dump. We are building infrastructure patterns.

  • No Scripts: Don't contribute a single file script. Contribute a system.
  • No Happy Paths: Every project must handle failure, rate limits, and bad inputs.
  • Ethics First: Every contribution must include an ETHICS.md discussing failure modes.

Project Structure

New projects must follow the strict template:

project-name/
├── README.md           # Problem statement, mental model
├── DESIGN.md           # Architecture, tradeoffs, diagrams
├── ETHICS.md           # Safety, bias, misuse
├── src/                # Implementation
├── tests/              # Unit and integration tests
├── evals/              # Golden datasets and scoring logic
└── traces/             # Example logs/traces

Standards

  1. Type Hints: All code must be strictly typed (mypy compliant).
  2. Schema Validation: Use Pydantic for all IO.
  3. Observability: All complex logic must be traced (OpenTelemetry or similar).
  4. Dependencies: Prefer raw clients (openai) over heavy frameoworks (langchain) unless the project is about the framework. We want to teach the mechanics.

Pull Request Process

  1. Open an issue proposing the Pattern.
  2. Receive approval on the architecture.
  3. Submit PR with full test coverage and an Eval run.