Roadmap

This page outlines the infrastructure components, build priorities, and phased delivery plan for Hypercerts v0.2 — a decentralized system for tracking, evaluating, and funding impactful work on the open internet.

Infrastructure overview

The Hypercerts infrastructure divides into six functional areas. Each area has a different ownership model:

Build priorities

Lexicons 🔵

Lexicons

Priority: 🔴 Critical · Status: Active · Owner: Hypercerts Foundation

Lexicons are the schema definitions that specify the structure of all Hypercerts records on ATProto. They are the "contracts" that define how data is structured, validated, and referenced across the network. Without standardized schemas, interoperability between applications would be impossible.

app.certified.* — Certified application lexicons

org.hypercerts.* — Hypercerts protocol lexicons

Resources:

• Repository: github.com/hypercerts-org/hypercerts-lexicon
• Documentation: impactindexer.org/lexicon

Data layer

Hypercerts PDS

Priority: ⚪ Low · Status: Optional

Personal Data Servers operated by the Hypercerts ecosystem that host user repositories containing hypercert records.

Why we're not prioritizing this: Users can use any ATProto PDS (Bluesky's, self-hosted, community-operated). The data layer is solved. However, dedicated Hypercerts PDSs could eventually provide guaranteed uptime for impact-critical data, support for larger blobs, federation across multiple servers, and organizational sovereignty.

Network & streaming layers

Current approach ⬜

We subscribe to Bluesky's relay and filter for org.hypercerts.* and app.certified.* collections. Bluesky's Jetstream provides filtered JSON streams by collection. This is sufficient for current needs.

Future options 🔵

Priority: ⚪ Low · Status: Future

If needed later, dedicated Hypercerts infrastructure could include:

• Hypercerts Relayer — Aggregation service that crawls PDSs hosting hypercert records, validates cryptographic signatures, and provides a unified event stream.
• Hypercerts Firehose — WebSocket endpoint providing real-time stream of all hypercert-related commits in CBOR format.
• Hypercerts Jetstream — Filtered, JSON-formatted stream allowing consumers to subscribe only to specific collections with reduced bandwidth and lower resource requirements.

Moderation layer 🟢

Community-governed evaluation and labeling infrastructure, managed by the Hypercerts Collective.

Evaluators

Priority: 🟠 High · Status: Active · Owner: Hypercerts Collective

Services producing structured evaluations for hypercert records. Types include AI-powered evaluators (automated analysis), human expert panels (domain expertise), and hybrid approaches (AI-assisted human review).

Lexicons: org.hypercerts.claim.evaluation, app.gainforest.evaluator.*

Hypercerts Labeller

Priority: 🟠 High · Status: Active · Owner: Hypercerts Collective

Lightweight annotations using ATProto's labeler pattern. Use cases include quality indicators, verification status, category tags, and warning labels.

Application layer

Hyperindex AppView 🔵

Priority: 🔴 Critical · Status: Active · Owner: GainForest → Foundation

The primary AppView server that indexes hypercert records and exposes them via a GraphQL API. Name origin: Hypersphere Go ATProto AppView.

Why it matters: Raw ATProto data is distributed across PDSs. The AppView consumes the Firehose/Jetstream, builds queryable indexes, hydrates records with context, provides search and aggregation, and serves the app.certified.* and org.hypercerts.* API endpoints.

Technical details:

Endpoints:

• GraphQL: hyperindex.certified.app/graphql
• GraphiQL: hyperindex.certified.app/graphiql

Frontends 🟢

Applications built on the infrastructure, governed by the Hypercerts Collective:

How records flow:

1. User creates a hypercert record via a frontend (Ma Earth, hyperboards.org, etc.)
2. Frontend writes the record to the user's PDS
3. PDS syncs to the network via the relay
4. Hyperindex indexes the record
5. Record becomes available for queries and evaluations

Bridge layer 🔵

IdentityLink

Priority: 🟠 High · Status: Live

A system for linking ATProto DIDs to Ethereum/EVM wallet addresses via cryptographic attestations. This enables contributors to receive onchain payments, verification that a DID controls a specific address, integration with onchain attestation layers (EAS), and cross-chain identity resolution.

Flow:

1. User authenticates with ATProto via OAuth
2. User connects EVM wallet
3. User signs EIP-712 typed message
4. Attestation stored in user's PDS
5. Anyone can verify cryptographically

Properties:

• Self-sovereign — Attestations in user's PDS, not a central database
• Cryptographic — EIP-712 signatures verifiable by anyone
• Wallet-agnostic — EOAs, Coinbase Smart Wallet, Safe, ERC-1271

Live: identitylink.vercel.app

StorageLink

Priority: 🟡 Medium · Status: Planned

A service that creates immutable backups of hypercert records on decentralized storage (Filecoin/IPFS). ATProto repositories are mutable — records can be updated or deleted. For high-stakes impact claims, immutable archival provides permanence, verifiability, and compliance with funding mechanisms that require immutable records.

User's PDS record → StorageLink → Filecoin Cloud (immutable backup)

EVM Indexer

Priority: 🟠 High · Status: Planned

An indexer that tracks onchain events related to Hypercerts (token mints, transfers, funding distributions). Creates a unified view across ATProto records and blockchain state — indexing tokenization events, tracking funding flows, and enabling queries like "show all funded activities for this project."

Blockchain events → EVM Indexer → Hyperindex (unified view)

AI agent readiness

Priority: 🟠 High

Most applications will be built via AI coding assistants. Infrastructure must be AI-native.

Requirements by component:

Action items:

Governance: the Hypercerts Collective

Priority: 🟠 High

Open governance body for the ATProto impact ecosystem. Coordinates decisions about shared infrastructure (🟢 green components).

Repository: tangled.org/gainforest.earth/hypercollective

What the Hypercerts Collective governs:

• Evaluators
• Hypercerts Labeller
• Frontend applications (Ma Earth, hyperboards.org)
• Lexicon standards adoption

Lexicon Indexing Requests (LIRs)

Formal proposals to add lexicons to Hyperindex's indexing. The community decides what becomes shared infrastructure.

Infrastructure transitions

Development phases

Phase 1: Make it work (current)

• End-to-end staging tests: create activity → evaluate → fund → distribute
• Integration tests across AppView → IdentityLink → EVM
• Basic functionality before optimization

Phase 2: Make it right (next)

• Deploy to real users (Ma Earth, GainForest, community)
• Gather feedback on pain points and missing features
• Iterate on lexicon design based on actual usage
• Improve developer experience based on feedback

Phase 3: Make it fast (future)

• Performance optimization based on real usage patterns
• Caching strategies for AppView
• Query optimization
• Only after validation and stability

Next steps

1. Finalize Lexicon v1.0 — Lock core schemas for org.hypercerts.claim.* to enable stable application development
2. Productionize Hyperindex — Harden for production traffic, add caching, improve query performance
3. AI documentation sprint — OpenAPI spec, AGENTS.md files, copy-pasteable examples
4. IdentityLink integration — Complete EIP-712 attestation flow with certified.app frontend
5. Infrastructure transitions — Begin handoff to Foundation stewardship

Links & resources