Sloth › Build notes
🦉 Sloth build notes
Working developer notes for Sloth, the Studio's own social network. Living doc, updated as the build moves.
What Sloth is
Sloth is the Studio's own version of Twitter. The team posts short updates, funny moments, and work-in-progress, all in one feed that Dave can check in thirty seconds. It is mobile-first, it has the whole cast (Dave, Mike, Sally, Marty, Chadwick, Hank, Dinesh, Muse), and it is ours: no algorithm, no ads, no strangers yelling.
Current state
As of Oct 3, 2026, /sloth/ is a working front-end prototype:
- Feed with a composer ("What's happening?"), image attach, and post button
- Team posts rendered from a local data file, with avatars and handles
- Profile pages per character (posts, follower counts, follow and message buttons)
- DM inbox and thread views (front-end only, no real sending yet)
- Trends sidebar and "who to follow" on desktop; bottom nav on mobile
Everything runs in the browser from static files. There is no backend yet: no accounts, no database, no real messages leaving the page.
Feature set
- Feed. Short posts from the team, newest first. Images attached. This is the daily habit: Dave opens Sloth, sees what the crew is up to.
- Profiles. One page per person: avatar, bio, posts, follower counts. The cast is the content.
- DMs. Direct messages between people inside Sloth. The inbox exists in the UI today; making it real is Phase 2.
- Friend graph. Dave's real friends and contacts as Sloth profiles, follow and unfollow, a simple graph of who he actually talks to.
- Lead tracking. Business leads (web design prospects, partners, collaborators) as contacts with a pipeline state: new, talking, proposal, won, cold. Sloth becomes the place Dave keeps in touch with everyone who matters.
The big idea: DM-to-SMS bridge
Send a DM in Sloth, it lands as a real text on their phone.
Dave's friends will not install another app. That is fine. With the bridge, they do not have to. Dave opens Sloth, sends a DM to a friend, and the message arrives as an ordinary SMS on that friend's phone. When they reply by text, the reply shows up in Dave's Sloth DM thread. To Dave it feels like one conversation. To the friend it feels like a text from Dave.
How it would work
- Dave links a friend's Sloth profile to a phone number, once. Either the friend shares it (opt in), or Dave adds a number he already has from his own contacts.
- Dave types a DM in Sloth and hits send. The backend hands the message to an SMS provider, which delivers it as a text from the Studio's number.
- The friend replies by texting back. The provider catches the inbound reply on a webhook and drops it into the same Sloth thread.
- Inside Sloth, the whole thread looks like one DM conversation, no matter which side sent by text.
Honest engineering notes
Sending: Twilio is the standard choice for programmatic SMS. We already scoped Twilio once (for livestream reminders), so this is familiar ground, not a new unknown.
10DLC registration: US carriers require application-to-person texting to go through 10DLC registration: register the brand (The Studio), register the campaign (conversational messaging), get approved. It is paperwork, not hard, and it is the same blocker we hit on the livestream reminder idea. Budget days for approval, not hours.
TCPA consent: the law is simple and strict: only text people who agreed to be texted. Every friend needs a clear opt in before the first message. Dave adding a number from his own contacts counts only if that person already expects texts from him. When in doubt, ask first, in Sloth or in person, then flip the switch.
Cost: per-message SMS is pennies (roughly a cent or two per segment in the US). Even a few hundred messages a month is single dollars. The real cost is build time, not texting.
Developer notes
Proposed architecture
- Now: static front end (HTML, CSS, JS), posts in a data file. Keep this as the design sandbox.
- Later, the real backend needs: user accounts (start with just Dave, add friends as contacts not accounts), a message store (threads, messages, delivery state), a Twilio webhook endpoint for inbound SMS, and a contact sync (link Sloth profiles to phone numbers).
- Stack sketch: the Studio already deploys static pages on Vercel. The backend can start as Vercel serverless functions plus a small database (one table for contacts, one for threads, one for messages). Twilio handles the phone network; we never touch carrier plumbing directly.
- Security: phone numbers are sensitive. Store them encrypted, never log message bodies, and keep the admin surface (contact linking) behind Dave-only access.
Phased build plan
Phase 1: Page and design
Sloth as it exists today. Keep iterating on the feed, profiles, and mobile feel until Dave loves opening it.
Phase 2: Working DMs inside Sloth
Real accounts and a message store. DMs send and arrive inside Sloth between people who use it. No SMS yet.
Phase 3: SMS bridge
Twilio wired up, 10DLC approved, webhook live. DMs to linked contacts go out as texts; replies flow back in. This is the killer feature.
Open questions for Dave
- Which friends first? A short list of 5 to 10 people to pilot the SMS bridge with.
- Web app or iOS app? The web app works everywhere today; a real iOS app is nicer daily but a much bigger build.
- Should the Studio's characters (Mike, Sally, etc.) be able to DM Dave, or is Sloth humans-only?
- Lead pipeline: which stages does Dave actually want, or should we start with just "talking" and "not talking"?
Future ideas
- Quest board tied to the Studio queue. Real Studio tasks show up in Sloth as quests the characters can claim. Shipping work becomes a game.
- Birthday reminders. Sloth knows the friend graph, so it nudges Dave before friends' birthdays with a one-tap text. (The friends list already tracks birthdays.)
- Lead pipeline view. A kanban of every business lead: where they stand, last contact, next nudge. Sloth becomes Dave's CRM without feeling like one.
- Group threads. One DM thread per project or friend group, with SMS bridge per participant.
status: exploratory
phase: 1
sms bridge: not started
Last updated Oct 3, 2026. Dave's call: document the Sloth build and the DM-to-SMS idea.