Product Requirements Document
Version: 2.0
Date: March 15, 2026
Status: Draft
Product: Tribe
Tribe is a mobile-first social product built around the idea that human relationships have natural limits. Instead of rewarding endless accumulation of followers, content, and noise, Tribe helps people intentionally organize and experience their real social world.
The core product uses relationship tiers inspired by Dunbar-style social group sizes. Users place people into meaningful circles with explicit limits, then choose how and where they share. The result is a calmer, more human social experience centered on closeness, context, and reciprocity rather than scale, virality, and algorithmic engagement.
Tribe is designed for people who feel alienated by traditional social apps but still want lightweight, visual, ambient connection with the people who matter most.
Build a social app that treats attention, intimacy, and social capacity as finite and worth protecting.
Tribe helps users:
- focus on the people they actually care about
- share with the right level of intimacy
- reduce social overload
- maintain healthier, more intentional digital relationships
- Finite by design: limits are product features, not bugs
- Relationship-aware: audience and context matter
- Portable and user-owned: identity and graph should not be trapped in a single platform
- Low-noise by default: no endless feeds, no engagement bait, no growth hacking mechanics
- Honest social UX: the app should reflect real social dynamics, not flatten everyone into the same “friend” bucket
Current social platforms optimize for volume, reach, and time spent. That creates a few recurring problems:
- users maintain far more weak ties than they can meaningfully engage with
- close relationships get buried under casual or performative posting
- audience collapse makes sharing feel awkward or emotionally risky
- users lose trust in feeds shaped by opaque algorithms
- “friend” and “follower” models do not reflect real human social structure
There is an opportunity to build a social experience that feels smaller, safer, and more intentional without feeling empty or overly restrictive.
People in their 20s to 40s who are socially online but emotionally fatigued by mainstream platforms.
-
Socially overloaded professionals
They want connection without a firehose of content. -
Thoughtful sharers
They like posting photos, life updates, and voice notes, but want tighter audience control. -
Community-oriented users
They care about real relationships and small groups more than public reach. -
Ex-social-power-users
People who are tired of Instagram, TikTok, X, or Facebook but still want a living social layer.
- “Show me posts from people I genuinely care about.”
- “Let me share differently with different circles.”
- “Help me keep my network sane.”
- “Do not reward me for posting to everyone all the time.”
- “Let me leave or move my identity without losing everything.”
- Create a distinctly calmer and more intentional social experience
- Increase the quality of interaction over quantity of interaction
- Make audience selection intuitive and habitual
- Prove that explicit relationship tiers can improve user satisfaction and retention
- Establish a differentiated position within the AT Protocol ecosystem
- Enable portability of graph, identity, and content
- Create a foundation for privacy-respecting social tools and community extensions
- Becoming a general-purpose creator platform
- Competing on public discovery or viral growth
- Supporting large-scale influencer monetization
- Replacing every existing messaging or social app on day one
Tribe organizes a user’s social world into bounded relationship layers. Each layer has a capacity limit and a distinct sharing context.
- Inner Circle: up to 5 people
- Close: up to 15 people
- Tribe: up to 50 people
- Village: up to 150 people
Users can view separate feeds by tier and choose the audience for each post. The app nudges users to maintain a realistic graph and periodically review inactive or drifting relationships.
This model gives users:
- a more accurate social map
- better audience control
- lower feed volume
- more meaningful defaults for sharing
Users can:
- sign up and authenticate using AT Protocol-compatible identity
- create a profile with display name, avatar, and short bio
- import or connect existing AT-based identity if applicable
- support DID-based identity model
- provide secure key management with clear recovery flows
- keep onboarding understandable for non-technical users
Users begin with empty tiers and are guided to populate them intentionally.
- create and manage tier assignments
- add people through search, contacts, suggestions, or manual entry
- move people between tiers
- enforce tier caps
- show remaining capacity per tier
- support a review step before finalizing initial graph
- onboarding should explain why limits exist
- limits should feel thoughtful, not punitive
- users should understand that tiers are adjustable over time
MVP includes separate feeds by relationship tier.
- Inner Circle
- Close
- Tribe
- Village
- chronological ordering only
- no infinite scroll
- limited session depth, such as pagination or natural stopping points
- clear context about why each post is visible
- text posts
- photos
- short captions
- lightweight reactions
- comments
- voice notes
- ephemeral posts for Inner Circle
Users choose a sharing audience when creating a post.
- composer supports tier-specific posting
- default audience is explicit, never hidden
- users can preview who will see a post
- audience options include:
- Inner Circle
- Close
- Tribe
- Village
- Broadcast
Broadcast exists in MVP only if simple enough to implement. It should be intentionally limited and visually distinct from normal sharing. The product should discourage overuse rather than optimize it.
The app should help users maintain a coherent social graph.
- Capacity alerts: notify users when a tier is nearly full
- Inactivity review: suggest reviewing relationships with no meaningful interaction after a defined period
- Tier review prompts: periodically ask whether a person still belongs in the same circle
Nudges should feel reflective and helpful, not shaming or manipulative.
Users need confidence that tiered sharing actually means something.
- control who can see each post based on selected audience
- block and mute users
- hide posts from specific users if needed
- basic content reporting and abuse escalation
- clear explanation of what is public, semi-private, and private
Tribe should use AT Protocol where it provides strategic value while avoiding unnecessary technical complexity in the first release.
- account identity compatible with AT Protocol
- user content and graph stored in a portable, standards-aligned way where practical
- support future migration between providers
- ensure architecture allows interoperability with broader ecosystem over time
- Bluesky <> Tribe graph or content syncing
- full plugin ecosystem
- complex federation features exposed directly in product
- advanced third-party custom feeds at launch
These are promising but should not block MVP.
An opt-in syncing layer that lets users carry more of their AT Protocol social world into Tribe and optionally project selected Tribe activity back out to Bluesky-compatible clients.
Potential directions:
- import Bluesky follows, followers, or curated lists as suggestions for Tribe tiers
- map selected Tribe relationships to app-private graph metadata without exposing private tier structure publicly
- optionally mirror eligible Tribe posts to the user's AT Protocol repo with clear audience and privacy constraints
- optionally surface selected public AT Protocol posts inside Tribe when they match the user's relationship context
- provide user controls for one-way import, one-way publish, or two-way sync
- make sync state understandable, reversible, and explicit at all times
An opt-in feature that lets users understand roughly how reciprocal a relationship is. This should be handled carefully because it can create emotional harm if presented bluntly.
A mechanism that limits high-volume broad sharing and encourages selectivity.
Users define personal content boundaries such as politics, conflict, or event spam.
Private stories, quick voice notes, or temporary check-ins for closest ties.
Small group albums, event planning, polls, and ritual-based social prompts.
Private, user-facing insights such as:
- who they engage with most
- which tiers are overloaded
- whether their graph is stable over time
The app should feel finite, breathable, and emotionally safe.
Every major action should make audience and relationship context obvious.
Adding, broadcasting, or overfilling a tier can have intentional friction. Reading, replying, and staying connected should feel easy.
The product can be honest without being cruel. Tone matters.
Users should feel that their identity and relationships belong to them, not the app.
- user authentication and profile setup
- tier creation and management with hard caps
- searchable user graph and adding people to tiers
- tier-based feeds
- post composer with audience selection
- chronological feed ranking
- comments and lightweight reactions
- block and mute controls
- inactivity and capacity nudges
- foundational AT Protocol-compatible architecture
- contact import
- voice notes
- ephemeral content
- shared albums
- group polls
- limited broadcast budgeting
- percentage of users who complete tier setup during onboarding
- average number of users placed into at least two tiers within first 7 days
- weekly active users who post or interact within Inner Circle or Close
- average feed session length compared to post-session satisfaction
- retention at day 7, day 30, and day 90
- percentage of posts shared to non-broadcast audiences
- average comments or reactions per post within close tiers
- frequency of tier adjustments over time
- percentage of users who report better signal-to-noise than on mainstream social apps
- user complaints about confusing privacy boundaries
- churn caused by feeling overly constrained
- abuse, harassment, or trust incidents
- onboarding drop-off caused by too much graph setup friction
- Client: React Native
- Backend / Protocol Layer: AT Protocol SDK and related services
- Identity: DID-based authentication and account portability model
- Storage: protocol-aligned user data repositories plus media storage strategy appropriate for privacy and performance
- Analytics: privacy-preserving product analytics with minimal collection
- secure identity and recovery
- audience enforcement and privacy correctness
- portable data model for tiers and posts
- low-latency feed rendering
- simple architecture for early iteration
- how tier metadata should be represented in a protocol-compatible way
- what content can be fully portable versus app-specific
- best approach for private or encrypted content within protocol constraints
- whether media should be stored directly in protocol-linked storage or a hybrid architecture
- how Bluesky follows, lists, and public posts should map into Tribe tiers without collapsing private relationship intent
- whether Tribe-to-Bluesky sync should be publish-only, import-only, or selectively bi-directional
- how sync conflicts, deletions, and privacy downgrades should be handled across Tribe and Bluesky surfaces
Mitigation: position limits as a benefit, teach the model clearly, and let users reassign people easily without social drama.
Mitigation: emphasize that tiers are private, flexible, and for user clarity, not public ranking.
Mitigation: make audience selection and visibility states extremely explicit throughout the product.
Mitigation: keep interoperability goals strong but narrow MVP scope to what can be built reliably.
Mitigation: optimize for warmth and relevance, not volume. Ensure initial onboarding quickly creates a living feed.
Mitigation: keep such features out of MVP or make them strictly opt-in with careful wording.
- finalize user flows
- design social tier model
- define data model and privacy rules
- prototype onboarding and posting UX
- authentication and identity
- tier management
- posting and feeds
- privacy controls
- core nudges
- recruit a small cohort of intentional early adopters
- observe onboarding success and social graph behavior
- test whether limits feel empowering or annoying
- validate trust in audience controls
- launch with strong product narrative around calmer social connection
- target users disillusioned with mainstream feeds
- position Tribe as a relationship-first network, not another content app
A relationship-first social app built around real human capacity.
Tribe is not “anti-social media.” It is pro-context, pro-intimacy, and pro-signal.
- explicit relationship tiers with meaningful limits
- bounded feeds instead of endless consumption
- audience-aware sharing as a first-class feature
- portability and protocol-aligned identity
- a product philosophy centered on human capacity rather than growth at all costs
- Should Broadcast exist in MVP or wait until the core tier model proves itself?
- Are tier names best kept as fixed defaults, or should users rename them?
- Should users be able to follow people outside their managed tiers in a read-only way?
- How much friction is useful when demoting or removing someone?
- What is the right balance between private intimacy and interoperability?
- Which features best create delight without undermining the core discipline of the product?
Tribe has a strong and differentiated idea at its core: social software should reflect the actual limits of human attention and closeness. The product opportunity is not to help users scale their network endlessly, but to help them experience their relationships more clearly and with less noise.
The MVP should stay focused. The winning first version is not the most radical or the most feature-rich. It is the one that makes tiered social connection feel intuitive, calming, trustworthy, and worth coming back to.