Skip to main content

Neo Codex OS

An AI Engineering Operating System for durable project continuity.

Neo preserves engineering context, understands project state, and guides developers through the software lifecycle so they can spend less time reconstructing context and more time building.

Project constitution

Neo's founding direction is documented in:

Contributor and project policies:

Install locally

From this repository:

py -m pip install -e .

Verify:

neo --version

Initialize a project

Inside any project:

neo init

Or pass a path:

neo init C:\Projects\sample-fastapi

Check readiness

neo doctor

Show current project state

neo status
neo status C:\Projects\sample-fastapi --verbose

Status reports where the project is, including explicit uncertainty and conflicts. It does not recommend what to do next.

Create a verified handoff

Record explicit human continuation intent for a fresh coding-agent thread:

neo handoff --next-step "Run the focused regression test."
neo handoff --question "What causes the failing test?"

At least one of --next-step or --question is required. Neo always shows the complete proposed HANDOFF.md and asks for explicit authorization before its protected one-file replacement. Repeating an equivalent handoff is a verified, non-mutating skip. These values are human statements, not Neo recommendations.

Print reusable prompts

neo prompt start
neo prompt handoff
neo prompt debug
neo prompt deploy

Generated project files

AGENTS.md
HANDOFF.md
TASKS.md
DECISIONS.md
.neo/
  context-authoring.md
  start-thread.md
  end-thread.md
  debug.md
  deploy.md

Existing files are preserved unless --force is used.

Before editing Schema 0.1 structured sections, read .neo/context-authoring.md. It documents the exact writable grammar without acting as another context document or ProjectState source.

Development

Install Neo with its bounded development toolchain:

py -m pip install -e ".[dev]"

Run the same quality gates used by CI:

py -m ruff format --check src tests
py -m ruff check src tests
py -m mypy
py -m pytest
py -m build
py -m twine check dist/*

Use py -m ruff format src tests to apply formatting. GitHub Actions runs the test suite on Windows and Linux across Python 3.10 through 3.14 and smoke-tests built distributions on both operating systems.

See CONTRIBUTING.md for the complete contribution workflow.

Typical workflow

cd C:\Projects\my-project
neo init
neo doctor
neo status
neo handoff --question "What should the next session investigate?"
neo prompt start

Official skills

Neo's first repository-owned skill is:

skills/
  neo-architect/
    SKILL.md
    agents/openai.yaml

neo-architect guides architecture and substantial software work through an adaptive SDLC, integrating with Neo context files when they exist. It is currently a repository asset; the neo CLI and Python package do not yet install skills into an agent environment.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

infinite_neo-0.1.0.tar.gz (120.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

infinite_neo-0.1.0-py3-none-any.whl (87.0 kB view details)

Uploaded Python 3

File details

Details for the file infinite_neo-0.1.0.tar.gz.

File metadata

  • Download URL: infinite_neo-0.1.0.tar.gz
  • Upload date:
  • Size: 120.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.0

File hashes

Hashes for infinite_neo-0.1.0.tar.gz
Algorithm Hash digest
SHA256 c90bac8a2e5bd030956a26a949eeaa7076d27d8caeab1888f494ea43b99e3829
MD5 da6b0577054adb05e7007a2e9f7d2eb5
BLAKE2b-256 367593bbabc0ee37f57a17d533a9128b08da8d371b0a293082eb15192c562c82

See more details on using hashes here.

File details

Details for the file infinite_neo-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: infinite_neo-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 87.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.0

File hashes

Hashes for infinite_neo-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 436abf2b6953162247bed077dc3015b4be8d2c9a457713e5ab4a99f915a10de1
MD5 a2bb39998c0fbf0fa924a739594766b8
BLAKE2b-256 45137acfa7ad31ba994e895b65596f7dc240e2335e40308c32cf15736ccf773b

See more details on using hashes here.

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page