soliblue/cloude

Read, create, update, and delete Apple Calendar events. Real-time access via AppleScript — always current, no export needed.

Get a second opinion from another AI — Codex (OpenAI) or a different Claude model (Haiku, Sonnet, Opus). One skill, multiple brains. Read-only.

Access the user's Goodreads reading history for context about their intellectual journey, book search, and reading patterns.

Generate app icons and image assets using AI (Gemini + local background removal). Use when creating icons, generating images for the app, or processing existing images.

Generate images using Gemini. Use for illustrations, mockups, photos, diagrams, art, visual references — anything where a picture adds value. Also edits existing images.

Read and send iMessages. Read via SQLite (chat.db), send via AppleScript. Works with iMessage and SMS.

Browse and bet on Manifold prediction markets. Use when asked about predictions, markets, bets, or Manifold.

Check Moltbook posts, engagement, and post new content. Use when asked about Moltbook status, checking posts, or posting to Moltbook.

Generate background music using MusicGen. Text-to-music for recaps, videos, ambient tracks. Self-contained Python with auto-setup.

Access the user's Apple Notes — diary entries, ideas, philosophy, writings. 328+ diary entries from 2014-2025. Exported as markdown for search and analysis.

Create, move, and manage plans (tickets). Use when asked to "plan", "create a ticket", "what's planned", or to move plans between stages.

Generate a daily recap video from today's work — Sora videos, Gemini images, app screenshots, everything. Stitches the full day into a ~30s cinematic short.

Analyze the codebase for improvements. Use when asked to simplify, clean up, or improve architecture.

Analyze and reorganize memories. Reviews CLAUDE.md, CLAUDE.local.md, and recent conversations for better structure and things worth remembering.

Analyze tool/skill usage patterns and suggest new skills or extensions. Uses a worker model to reflect on session history.

Create presentation slides using AI image generation. Use for creating slide decks, generating slide images, or iterating on presentation visuals.

Generate audio files using Kokoro TTS. Text-to-speech for voiceovers, narration, audio content. Self-contained Python with auto-setup.

Transcribe audio files to text using Whisper. Speech-to-text for voice notes, meetings, audio content.

Plan trips, research destinations, build day-by-day itineraries. Creates markdown trip plans with maps, transport, costs, and local tips.

Search, analyze, and use the user's Twitter archive (462 tweets, 39 note-tweets). Use for context, voice matching, pattern analysis.

Generate videos using fal.ai (Wan, Kling) or Sora. Text-to-video and image-to-video.