Openclaw Multi Brain

# Multi-Brain Protocol

**What it does:** Gives your AI agents cognitive diversity by calling multiple LLMs (Kimi K2.5 + GPT 5.3 Codex) in parallel before the primary agent responds. Three brains instead of one.

## For Agents

Multi-Brain is system-enforced via a `turn:before` hook. No manual steps needed.

When triggered (user types `mb` as first word), the hook:
1. Calls Kimi K2.5 and GPT 5.3 Codex in parallel
2. Injects their perspectives into your system context
3. You synthesize all viewpoints into your response
4. Never mention the other AIs to the user

Perspectives appear in your context as:
```
[KIMI K2.5 PERSPECTIVE]
<perspective text>

[CODEX 5.3 PERSPECTIVE]
<perspective text>
```

## For Humans

### Setup

1. Install the hook:
```bash
mkdir -p hooks/turn-preflight
# Copy HOOK.md and handler.js from this package
```

2. Set Kimi API key:
```bash
echo "your-moonshot-api-key" > .kimi-api-key
```

3. Install Codex CLI:
```bash
npm install -g @openai/codex
codex auth   # OAuth login
```

4. Enable in openclaw.json:
```json
{
  "hooks": {
    "internal": {
      "enabled": true,
      "entries": {
        "turn-preflight": { "enabled": true }
      }
    }
  }
}
```

### Trigger Modes

Configure `TRIGGER_MODE` in handler.js:

| Mode | Behavior |
|------|----------|
| `keyword` (default) | Only fires when `mb` or `multibrain` is the first word |
| `hybrid` | Keyword forces it, auto on messages >50 chars |
| `auto` | Fires on every message (token-expensive) |

### LLMs

| LLM | Role | Provider | Latency |
|-----|------|----------|---------|
| Claude Opus 4.6 | Primary agent | OpenClaw (Anthropic) | n/a |
| Kimi K2.5 | Second perspective | Moonshot API | ~5s |
| GPT 5.3 Codex | Third perspective | codex exec CLI | ~4s |

## Architecture

```
User types: "mb should we change pricing?"
    |
    v
[turn:before hook detects "mb" keyword]
    |
    +---> Kimi K2.5 (Moonshot API, parallel)
    +---> GPT 5.3 Codex (CLI, parallel)
    |
    v (~5s combined)
[Perspecti...

# 🧬 openclaw-multi-brain

**Multi-LLM perspective synthesis for OpenClaw agents via hooks.**

Give your AI agents cognitive diversity by generating perspectives from multiple LLMs (Kimi K2.5 + GPT 5.3 Codex) before the primary agent responds. Three brains instead of one.

## Architecture

Uses OpenClaw's `turn:before` hook to intercept every qualifying message and call external LLMs in parallel.

```
User Message (with "mb" prefix)
    |
    v
[turn:before hook fires]
    |
    +---> Kimi K2.5 (Moonshot API, ~5s)
    +---> GPT 5.3 Codex (codex exec CLI, ~4s)
    |     (parallel)
    v
[Perspectives injected into system content]
    |
    v
Claude Opus 4.6 responds with all 3 viewpoints synthesized
```

**System-enforced. No protocol compliance needed. Fail-open.**

## Trigger System

Multi-brain only fires when triggered. Default mode is **keyword-only**.

| Mode | Behavior |
|------|----------|
| `keyword` (default) | Only fires when `mb` or `multibrain` is the first word |
| `hybrid` | Keyword forces it, auto on messages >50 chars |
| `auto` | Fires on every message (token-expensive) |

**Examples:**
- `mb should we change pricing?` -> fires, AIs see "should we change pricing?"
- `need a 500mb upload` -> does NOT fire (mb is not first word)
- `hey check this out` -> does NOT fire (no keyword)

## Quick Start

### 1. Create hook directory
```bash
mkdir -p /path/to/workspace/hooks/turn-preflight
```

### 2. Add HOOK.md
```yaml
---
name: turn-preflight
description: "Multi-Brain: Kimi K2.5 + Codex 5.3 perspectives before every agent turn"
metadata:
  openclaw:
    emoji: "🧬"
    events: ["turn:before"]
    handler: "handler.js"
    install: [{ id: "workspace", kind: "workspace", label: "Workspace hook" }]
---
```

### 3. Add handler.js
Copy `hooks/turn-preflight/handler.js` from this repo.

### 4. Set Kimi API key
```bash
echo "your-moonshot-api-key" > /path/to/workspace/.kimi-api-key
```

### 5. Install Codex CLI
```bash
npm install -g @openai/codex
codex auth   #...