Questions, answered.

Yeah, and we're not ashamed of it. We run Claude Opus and Sonnet (Anthropic) for reasoning and text, plus media routes like GPT Image 2, Nano Banana 2, Z-Image, Grok, Veo 3.1, Seedance 2.0, Kling, and Topaz. Frontier labs spend billions training those models. What we do is make them usable for specific jobs: hand-written skill files, mode-specific prompts, exact pricing, and real output filters.

One big AI that does everything falls back on the average. A résumé specialist has spent its whole existence thinking about résumés. Specialists read the room; generalists just finish the sentence.

Merl uses Claude Opus and Sonnet (Anthropic) for reasoning, writing, agent skills, and prompt refinement. Media routes include GPT Image 2, Nano Banana 2, Z-Image, and Grok for image work, plus Grok, Veo 3.1 4K, Seedance 2.0, Kling 3.0, and Topaz for video, face swap, and enhance. Raw model access is half the value at best. Merl layers prompt-craft, skill files, and output filters on top.

Each agent has a hand-written skill file — hundreds of lines of instructions, examples, counter-examples, failure modes, and taste rules. Editor's skill file knows what a startup PM résumé looks like different from an enterprise one. Echo's skill file knows how to preserve voice without flattening it. It's the difference between giving ChatGPT a one-line prompt and giving a specialist ten years of taste. We write these skill files, test them, and update them — you just pick the agent.

A second pass on every generation. Shine runs face-identity checks and rejects drifted results. Cutter checks video output (Seedance 2.0, Kling, Veo 3.1 — whichever model it routed to) for face warping and motion smear. Editor flags invented experience before it reaches you. ~30–90% of raw model output gets rejected and re-rolled before you see the final result. You only ever see the cuts that pass.

Editor (résumés + cover letters) · Echo (writing in your voice) · Shine (photo studio for people, products, objects, and styles) · Cutter (AI video, face swap, and enhance via fal.ai) · Designer (interior redesign) · Lore (private iMessage Wrapped and local chat-history answers).

Lore turns your imported Messages history into private, local iMessage Wrapped-style cards: full Wrapped decks, one-person wraps, date-specific mini Wrapped pages, and one-page answers. You import a copy of your Mac chat.db; Lore builds a local index on your device and ignores photos, videos, audio, and attachments.

Not in the public launch build. Merl is iOS-first right now: iPhone and iPad through the App Store.

Not for the public launch. The current product users can install is the iOS app.

Yes. More highly-trained specialists are in the pipeline — but we only ship an agent when it hits the same quality bar as the existing crew. For now, the crew is Editor, Echo, Shine, Cutter, Designer, and Lore.

Yes. One purchase, both iPhone and iPad. Same App Store listing.

New users start with 25 mana. Paid plans: Day Pass $4.99 · Weekly $9.99 · Pro $19.99/month. See pricing for mana breakdowns.

Text casts are low fixed mana. Lore answers are 5 mana, date-specific mini Wrapped pages are 35 mana, one-person wraps are 75 mana, and full iMessage Wrapped is 125 mana. Image casts start at 2 mana and scale by selected model/quality. Cutter prices exactly from the model, seconds, audio, and quality you choose — from cheap Grok drafts to Veo 3.1 4K, Seedance 2.0, Kling Face Swap, and Topaz Enhance.

Top-up mana never expires. Subscription mana resets each billing cycle — 125 on Day Pass, 300 weekly, 700 monthly, or 1,500 on Creator — and unused subscription mana does not roll over.

Yes — through your Apple ID subscriptions. Cancel whenever, keep remaining mana for that period. No "hit us on email" traps.

No. Ever. Your inputs are not used to train Merl, Claude, fal.ai-routed media models, or anyone else's models. Normal scrolls can sync to your private account when signed in. Lore's imported chat.db, local index, cached Wraps, and local answers stay on your device.

Normal cast history is stored on-device and, when you sign in with Apple, synced to your private Merl account so it follows you across devices. Generated media and Lore's imported message index stay device-only. Photos/source media pass only to the selected provider for the cast and are not stored by us. See privacy for details.

No. Lore keeps the imported chat.db, local index, cached Wraps, and local answers on your device. When you ask Lore for a Wrap or answer, it sends only selected message-derived excerpts or aggregate stats needed for that request to Claude. It ignores photos, videos, audio, and attachments.

Yes, within each underlying model's license. Merl is built for work, ads, profiles, listings, client drafts, and creator content.

Yes. Paste the job description and your existing résumé. Editor extracts what the role actually asks for (not just keywords), rewrites your bullets to mirror the posting's vocabulary, quantifies every line using your real experience, and flags anything that would require inventing experience you don't have.

Yes. Paste 3–5 samples of your natural writing once; Echo builds a persistent voice profile from cadence, vocabulary, sentence length, and idiosyncrasies. Uses that profile for every cast — not generic "AI writing."

Yes. Shine is tuned specifically to preserve facial features, skin tone, and bone structure. A face-identity filter rejects any drifted result automatically — ~90% of raw generations get thrown out before you see them.

Cutter routes each shot through fal.ai to the right workflow: Grok for fast drafts, Veo 3.1 for cinematic/audio, Seedance 2.0 for premium motion, Kling 3.0 for face/motion workflows, and Topaz for upscaling. Merl's quality layer checks for face integrity and motion coherence before you see the result.

Yes. Designer extracts your room's architecture first — walls, windows, light direction, existing floor plan — before restyling. You get new furniture, palette, and textures placed where they actually fit your specific room. Not a generic room that vaguely matches the style.