Nexis

Nexis is a private autonomous research agent — and the foundation of a new primitive: a private data access layer for the agentic internet.

Every AI agent, trader, researcher, and journalist that queries the internet today leaves a trail. RPC providers log every wallet lookup. Search engines profile query patterns. Data providers like Nansen and Dune see your entire research strategy. The infrastructure you depend on is watching everything you do.

Nexis routes every query through Gensyn's AXL peer-to-peer privacy network, encrypts every result with AES-256, and stores it on 0G's decentralized storage with a key that never leaves your machine. Data providers see AXL. Storage nodes see encrypted noise. Nobody sees you. It researches community sentiment, onchain wallets across 7 chains, competitor strategies, market landscapes, and fund flows — all without leaving a trace. KeeperHub makes it fully autonomous: onchain events trigger private research automatically, and Nexis creates monitoring workflows programmatically without human intervention.

As AI agents become the primary consumers of internet data, the need for a privacy layer becomes critical. A trading agent that queries price feeds reveals its strategy. A research agent querying competitors reveals its roadmap. An investigative agent analyzing wallets tips off its subjects.

Nexis is the answer. Not a tool. Infrastructure.

Nexis is built in TypeScript on Node.js/Express, deployed on a GCP VPS with PM2. Three integrations form the core architecture:

Gensyn AXL — Every single outbound HTTP request is routed through the AXL P2P encrypted mesh network. Not selected requests — all of them. RPC calls, Reddit scraping, GitHub API, Etherscan queries, SearchAPI — everything goes through AXL. The binary spawns automatically on startup. Every API response includes routedViaAXL: true as proof.

0G Storage — Every research session is AES-256 encrypted locally before upload. The encryption key lives in SQLite and is never transmitted. Results are stored on 0G Galileo testnet with root hash anchored on-chain. 0G also powers cross-session learning — Nexis retrieves past sessions on similar topics, decrypts locally, and compares findings across runs to detect confirmations and contradictions.

KeeperHub — Bidirectional integration. KeeperHub monitors the blockchain and webhooks Nexis when events fire (whale movements, DeFi deposits, scheduled triggers). Nexis also programmatically creates KeeperHub workflows via API — a user saying "monitor this wallet" results in Nexis designing and deploying a full 5-node workflow automatically. An LLM call generates arbitrary complex workflows from natural language. 5 live production workflows running now.

The research layer uses an LLM planner (Claude Opus via TokenRouter) that maps natural language goals to capability chains. Community research runs dual-layer extraction (behavioral vs technical) with semantic clustering, three-tier signal validation, and evidence anchoring. Onchain intelligence uses Etherscan V2 multichain API across 7 chains with scam detection heuristics. All capabilities route through the AXL privacy layer.