NVIDIA-labs OO Agents (NOOA) is a Pythonic, model-agnostic framework that treats AI agents as native Python objects, integrating state, prompts, tools, and typed interfaces through a single class.

Stars

436

7-day growth

No data

Forks

51

Open issues

3

License

NOASSERTION

Last updated

2026-07-28

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It merges object-oriented design with LLM orchestration, allowing developers to define agent behavior using familiar Python constructs (methods, docstrings, type annotations) and built-in typed I/O, auto-retry, and tracing, reducing the need for separate tool schemas.

Who it is for

  • Python developers building AI agents
  • Researchers exploring agent orchestration patterns
  • Teams needing hybrid deterministic-LLM workflows
  • Developers seeking to reduce boilerplate in agent frameworks

Use cases

  • Customer support ticket triage with typed structured outputs
  • Customer feedback sentiment and topic analysis
  • Multi-step agent workflows with method chaining and state
  • Rapid prototyping of agent pipelines with minimal configuration

Strengths

  • Pythonic interface removes the need for separate tool definitions—methods and annotations are the contract
  • Typed contracts with auto-retry improve reliability and runtime safety
  • Supports both LLM-driven (via `...` bodies) and deterministic Python methods in the same class
  • Built-in tracing, optional CLI, memory, and evaluation subpackages enable debugging and testing

Considerations

  • Research software with rough edges; not production-ready yet
  • LLM-generated code may take dangerous actions, requiring OS-level sandboxing (e.g., containers, NVIDIA OpenShell)
  • In-process AST validators are defense-in-depth, not a containment boundary—can be bypassed by reflective Python

README quick start

Installation

Description

labs-OO-agents

Related repositories

Similar projects matched by category, topics, and programming language.

lopopolo
Featured
lopopolo GitHub avatar

harness-engineering

Harness Engineering is a methodology for improving coding agent outputs by carefully crafting the environment around them—providing curated context, tools, and executable constraints that encode an organization’s nonfunctional requirements and cumulative lessons.

AI & Machine LearningAI Agents
2,390
slvDev
Featured
slvDev GitHub avatar

esp32-ai

A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.

AI & Machine LearningLarge Language Models
1,960
littledivy
Featured
littledivy GitHub avatar

mimic

mimic captures traffic from any iOS or web app and automatically generates a Python client library that lets you call the app's API like a regular library.

AI & Machine Learning
1,482