piddi-os 0.1.0

Deterministic, explainable context optimization for LLM applications.

ContextOS

ContextOS — The right context. Every time.

ContextOS is an opinionated, open-source Python library for Context Engineering: building, selecting, and preparing high-quality context to include in prompts for Large Language Models (LLMs). It is model-agnostic and designed to work with hosted and local models.

Current release: v0.1.0

Highlights

• Opinionated pipeline for deterministic context selection
• Budget-aware token estimation and context trimming
• Pluggable ranking and optimization stages
• Strong type coverage (py.typed) and CI-tested code

Installation

Install the latest release from PyPI:

pip install piddi-os

Import the library:

from contextos import ContextEngine

Quick Start

from contextos import ContextEngine

engine = ContextEngine()
engine.add_text("Angular introduced Signals.")
engine.add_text("React introduced Hooks.")

prompt = engine.build(query="Explain Angular Signals", max_tokens=500)
print(prompt)

Features

• ContextEngine: simple, composable public API
• ContextStore: in-memory context management
• Pipeline stages: ranking, budgeting, optimization
• Token estimation and budget enforcement
• Explainable optimization and tracing

Examples & Docs

See the examples/ directory and the docs/ folder for usage patterns, architecture notes, and recipes.

License

This project is licensed under the MIT License. See the LICENSE file for details.

Maintainers

Maintained by the Keshav Labs.