Initial commit: Ajarbot with optimizations

Features:
- Multi-platform bot (Slack, Telegram)
- Memory system with SQLite FTS
- Tool use capabilities (file ops, commands)
- Scheduled tasks system
- Dynamic model switching (/sonnet, /haiku)
- Prompt caching for cost optimization

Optimizations:
- Default to Haiku 4.5 (12x cheaper)
- Reduced context: 3 messages, 2 memory results
- Optimized SOUL.md (48% smaller)
- Automatic caching when using Sonnet (90% savings)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

SETUP.md

@@ -0,0 +1,59 @@
+# Ajarbot Setup Guide
+
+## Quick Start
+
+1. **Clone the repository**
+   ```bash
+   git clone https://vulcan.apophisnetworking.net/jramos/ajarbot.git
+   cd ajarbot
+   ```
+
+2. **Set up Python environment**
+   ```bash
+   python -m venv venv
+   venv\Scripts\activate  # Windows
+   pip install -r requirements.txt
+   ```
+
+3. **Configure credentials**
+   ```bash
+   # Copy example files
+   copy .env.example .env
+   copy config\scheduled_tasks.example.yaml config\scheduled_tasks.yaml
+   copy config\adapters.yaml config\adapters.local.yaml
+   ```
+
+4. **Add your API keys**
+   - Edit `.env` and add your `ANTHROPIC_API_KEY`
+   - Edit `config\adapters.local.yaml` with your Slack/Telegram tokens
+   - Edit `config\scheduled_tasks.yaml` with your user/channel IDs
+
+5. **Run the bot**
+   ```bash
+   python bot_runner.py
+   ```
+
+## Important Files (NOT in Git)
+
+These files contain your secrets and are ignored by git:
+- `.env` - Your API keys
+- `config/adapters.local.yaml` - Your bot tokens
+- `config/scheduled_tasks.yaml` - Your user IDs
+- `memory_workspace/memory_index.db` - Your conversation history
+- `memory_workspace/memory/*.md` - Your daily logs
+
+## Model Switching Commands
+
+Send these commands to your bot:
+- `/haiku` - Switch to Haiku (cheap, fast)
+- `/sonnet` - Switch to Sonnet (smart, caching enabled)
+- `/status` - Check current model and settings
+
+## Cost Optimization
+
+- Default model: Haiku 4.5 (12x cheaper than Sonnet)
+- Prompt caching: Automatic when using Sonnet (90% savings)
+- Context optimized: 3 messages, 2 memory results
+- Max tool iterations: 5
+
+See [README.md](README.md) for full documentation.