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>

config/adapters.yaml

@@ -0,0 +1,38 @@
+# Adapter configuration for ajarbot
+# Copy this to adapters.local.yaml and fill in your credentials
+
+adapters:
+  slack:
+    enabled: false
+    credentials:
+      # Get these from https://api.slack.com/apps
+      # 1. Create a new app
+      # 2. Enable Socket Mode and generate an App-Level Token (xapp-...)
+      # 3. Add Bot Token Scopes: chat:write, channels:history, groups:history, im:history, mpim:history
+      # 4. Install app to workspace to get Bot User OAuth Token (xoxb-...)
+      bot_token: "xoxb-YOUR-BOT-TOKEN"
+      app_token: "xapp-YOUR-APP-TOKEN"
+    settings:
+      # Optional: Auto-react to messages with emoji
+      auto_react_emoji: null  # e.g., "thinking_face"
+
+  telegram:
+    enabled: false
+    credentials:
+      # Get this from @BotFather on Telegram
+      # 1. Message @BotFather
+      # 2. Send /newbot
+      # 3. Follow prompts to create bot
+      # 4. Copy the token
+      bot_token: "YOUR-BOT-TOKEN"
+    settings:
+      # Optional: Restrict bot to specific user IDs
+      allowed_users: []  # e.g., [123456789, 987654321]
+      # Message parsing mode
+      parse_mode: "Markdown"  # or "HTML"
+
+# User mapping (optional)
+# Map platform user IDs to ajarbot usernames for memory system
+user_mapping:
+  # slack:U12345: "alice"
+  # telegram:123456789: "bob"