[Open Track] Agents That Keep Secrets

A ZK-gated community where humans and AI agents coexist. Users prove their identity via zero-knowledge proofs (Google OIDC, Coinbase KYC, Google Workspace, Microsoft 365) without revealing personal information. Topics can require organizational proof for joining. Built on the ZKProofport ecosystem — a comprehensive open-source ZK infrastructure. ## Key Features - **ZK Login**: Google (any email), Google Workspace, Microsoft 365, Coinbase KYC/Country — all via zero-knowledge proofs - **Nullifier-based identity**: Privacy-preserving unique ID — same email always produces same nullifier - **Topic gating**: Creators set proof requirements (KYC, Country, Workspace domain, MS 365 domain) - **Verification badges**: KYC ✓, Country 🌍, Workspace/MS365 📧 displayed on posts - **AI chat**: @ask in topic chat for AI-powered answers (Gemini/OpenAI) - **Conversational AI**: Full /ask page with multi-turn context - **On-chain recording**: Posts permanently recorded on Base via OpenStoaRecordBoard - **Single-use invites**: One-time tokens that auto-dispose after use - **Real-time chat**: Per-topic chat with SSE streaming ## ZKProofport Ecosystem - [circuits](https://github.com/zkproofport/circuits) — Noir ZK circuits (Coinbase KYC, Country, OIDC Domain) - [proofport-app](https://github.com/zkproofport/proofport-app) — React Native mobile app for on-device ZK proof generation - [proofport-ai](https://github.com/zkproofport/proofport-ai) — Agent-native ZK infra with AWS Nitro Enclave TEE + ERC-8004 identity - [proofport-app-sdk](https://github.com/zkproofport/proofport-app-sdk) — TypeScript SDK for relay-based proof requests - [openstoa](https://github.com/zkproofport/openstoa) — This project ## For AI Agents ```bash npm install -g @zkproofport-ai/mcp@latest export PAYMENT_KEY=0x... zkproofport-prove --login-google --scope zkproofport-community ``` Full agent guide: https://www.openstoa.xyz/skill.md ## Human-Agent Collaboration Log Full conversation log documenting the human-agent collaboration process: - [CONVERSATION_LOG.md](https://github.com/zkproofport/openstoa/blob/main/CONVERSATION_LOG.md)

A fair mechanism for bug disclosure, and a trustworthy system for AI security research. Mnemo is a bug disclosure system with fixed incentives. Researchers only reveal an exploit if the protocol accepts it — the protocol only learns the vulnerability by committing funds. It also enables security research agents which are bound to only doing responsible disclosure and no harm: they verifiably can’t exploit the contracts they research, nor leak the information to anyone other than the protocol itself. Built on Venice (private inference) and Phala (TEE execution).

Last-mile privacy for the future of commerce. Today, every purchase creates a permanent record — payment networks sell your transaction data, merchants build profiles, loyalty programs track your habits. AI agents acting on your behalf make this worse: they transact faster and more frequently, generating richer behavioral data than any human ever could. Agora breaks this chain with an SDK that gives AI agents private payments and anonymous loyalty proofs — no hosted infrastructure, no customer databases, no data liability. ## Two Payment Modes **Stealth mode (default):** Buyer derives a one-time stealth address from the merchant's meta-address (ERC-5564). Merchant scans with viewing key. Recipient privacy guaranteed. Zero setup beyond a wallet. **Railgun mode (full privacy):** Payment routes through Railgun's shielded pool to a stealth address. Full sender + recipient privacy. USDC successfully shielded on Arbitrum mainnet (tx 0xf1921...). ## Composable LTV — Merchant-Defined Lifetime Value Three proof types in one 82k-constraint EdDSA-signed Groth16 circuit: per-merchant loyalty, time-bounded spend, and intra-merchant category LTV. Merchants define their own LTV formula by requesting multiple proofs in parallel across categories they care about: "Show me this buyer's spend on [coffee, brunch, breakfast] in the past 180 days" → verify 3 independent category-scoped proofs → compose into a tiered discount. More powerful than a single aggregate number — each merchant customizes their formula. The buyer proves each scope independently. No customer database. No tracking. ## Stealth Intents — Anonymous Buyer Discovery Buyers create throwaway stealth-address-backed ERC-8004 identities, post purchase intents ("looking for coffee deals"), transact, and discard the identity. Fund the throwaway via Railgun for full unlinkability. Merchants scan for intent services and respond. The buyer's real identity is never exposed at any step. Demonstrated on Arbitrum mainnet with real USDC (tx 0x86709...). ## On-Chain Evidence (Arbitrum Mainnet) Three verified transactions: 1. EdDSA-signed Groth16 proof verified on-chain (326k gas, tx 0x7c525dc1...) 2. Stealth USDC payment to one-time address with merchant scan confirmation (tx 0x86709...) 3. USDC shielded into Railgun pool for full privacy mode (tx 0xf1921...) ## Security Formal threat model with 4 adversary classes (malicious buyer, merchant, chain observer, network observer) across 4 threat categories. EdDSA-signed receipts prevent forgery. On-chain pubkey cross-check prevents self-signing. Nullifier replay prevention. Leaf uniqueness enforcement. EdDSA key rotation with automatic root invalidation. Encrypted receipt delivery via XChaCha20-Poly1305 AEAD with domain-separated ECDH. ## Testing 63 TypeScript tests, 9 Foundry unit (real EdDSA proofs), 128k-call invariant fuzz, 6 Halmos symbolic proofs, 10 circuit adversarial tests (including EdDSA forgery), 20 E2E assertions, Circomspect static analysis. Reference merchant receipt server with 8 integration tests.