Metadata-Version: 2.4
Name: alpha-omega
Version: 0.3.0
Summary: Dual-brain debate protocol — Claude (Alpha) vs Codex (Omega)
Author: Yegor Bolshakov
License: MIT License
        
        Copyright (c) 2026 Yegor Bolshakov
        
        Permission is hereby granted, free of charge, to any person obtaining a copy
        of this software and associated documentation files (the "Software"), to deal
        in the Software without restriction, including without limitation the rights
        to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
        copies of the Software, and to permit persons to whom the Software is
        furnished to do so, subject to the following conditions:
        
        The above copyright notice and this permission notice shall be included in all
        copies or substantial portions of the Software.
        
        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
        SOFTWARE.
        
Project-URL: Repository, https://github.com/xizzzy/alpha-omega
Keywords: ai,debate,multi-agent,claude,codex,dual-brain
Classifier: Development Status :: 3 - Alpha
Classifier: Environment :: Console
Classifier: Intended Audience :: Developers
Classifier: Topic :: Software Development :: Quality Assurance
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Dynamic: license-file

# Alpha-Omega

Dual-brain thinking tool. Two genuinely different AI systems (Claude + Codex) independently analyze a problem, then debate to find each other's blind spots.

**Not two opinions. One solution where blind spots are already covered.**

## Install

```bash
pipx install alpha-omega    # recommended
pip install alpha-omega      # or with pip
```

## Quick start

```bash
ao setup                     # check prerequisites + initialize project
ao debate "your question"    # full dual-brain debate
```

## Commands

| Command | What it does |
|---------|-------------|
| `ao setup` | First-time setup: prerequisites + init + skill install |
| `ao doctor` | Check Claude CLI, Codex CLI, auth, project state |
| `ao debate "question"` | Full debate: blind memos → critique → Sigma resolution |
| `ao review` | Quick code review on unstaged changes |
| `ao review --staged` | Review staged changes |
| `ao review --branch main` | Review all changes since branch |
| `ao implement <id>` | Execute a debate resolution with either brain |
| `ao recall <query>` | Search past decisions and reviews |
| `ao contradictions` | Find conflicting past decisions |
| `ao init` | Create .alpha-omega/ in current project |
| `ao history` | Show recent debate outcomes |
| `ao status` | Show project AO state |

## How it works

```
    ┌──────────┐                              ┌──────────┐
    │  Alpha   │          BLIND PHASE         │  Omega   │
    │  Claude  │  ── independently analyze ── │  Codex   │
    └────┬─────┘                              └────┬─────┘
         │              CRITIQUE PHASE              │
         │  ── steelman + critique + concessions ── │
         └──────────────────┬───────────────────────┘
                            │
                   ┌────────▼────────┐
                   │  Design Sigma   │  ← deterministic, no LLM
                   │  (resolution)   │
                   └────────┬────────┘
                            │
                   ┌────────▼────────┐
                   │  Artifact Pack  │  ← resolution + dissent + risks
                   └─────────────────┘
```

## Resolution states

| State | Meaning |
|-------|---------|
| `ADOPT` | Strong consensus |
| `ADOPT_WITH_DISSENT` | Winner with recorded minority concern |
| `RUN_EXPERIMENT` | Both plausible, need data |
| `NEEDS_USER_CHOICE` | Depends on user priorities |
| `INSUFFICIENT_EVIDENCE` | Both brains uncertain |
| `DEADLOCK` | Fundamental disagreement |

## Configuration

Per-project config in `.alpha-omega/config.json`:

```json
{
  "alpha_model": "claude-sonnet-4-5",
  "alpha_timeout": 300,
  "omega_timeout": 600,
  "review_timeout": 180,
  "implement_timeout": 900,
  "implement_max_turns": 6
}
```

CLI flags override config values.

## Requirements

- [Claude Code CLI](https://docs.anthropic.com/en/docs/claude-code) (`claude` in PATH)
- [OpenAI Codex CLI](https://github.com/openai/codex) (`codex` in PATH)
- Python 3.9+

Run `ao doctor` to verify all prerequisites.

## Project structure

```
Your project/
├── .alpha-omega/
│   ├── config.json      # Model and timeout settings
│   ├── decisions.md     # Decision log
│   ├── sessions/        # Structured JSON per debate
│   ├── debates/         # Full markdown transcripts
│   └── reviews/         # Review results
├── AGENTS.md            # Omega's project memory
└── CLAUDE.md            # Alpha's project memory
```

## Integration with Claude Code

`ao setup` installs a global skill. Use `/alpha-omega` from any Claude Code session.

## Philosophy

See [MANIFESTO.md](MANIFESTO.md) — on the necessary architecture of thought that cannot be monological.

## License

MIT
