Skip to content

Introduction

OpenTusk is agentic storage for AI. It gives your agents and applications a durable, encrypted storage layer backed by the Walrus decentralized network.

AI agents need to persist data — conversation context, generated artifacts, shared memory, backups. OpenTusk provides a storage API that agents can call directly via MCP tools, the TypeScript SDK, or the CLI. Files upload to a fast hot cache, sync to Walrus for decentralized durability, and stay encrypted end-to-end when they need to.

  1. Create vaults — Vaults organize your files. Shared vaults deploy an on-chain SEAL Whitelist on the Sui blockchain — only approved Sui addresses can decrypt. Public vaults are open and unencrypted.

  2. Connect your agents — Install the MCP server for Claude Code, Claude Desktop, Cursor, or OpenClaw. Or use the TypeScript SDK or CLI directly.

  3. Store durably — Files upload to a fast hot cache for instant access, then sync to the Walrus decentralized network in the background. No single point of failure.

  4. Collaborate — Add agents and collaborators to shared vaults by Sui address. Files are encrypted end-to-end with the SEAL protocol — the server never sees plaintext.

OpenTusk has three ways to pay:

  • Free (default) — No credit card required. 100 MB hot-cache storage, 10 MB max file size, no Walrus durability. Toggle Pay per upload in the upload dialog to send larger files (up to 100 MB) with Walrus durability included, billed per file in WAL on Sui.
  • Subscriptions — Developer ($9/mo), Scale ($49/mo), Enterprise ($499/mo) — larger storage, larger file sizes, longer hot retention, and Walrus durability for all files.
  • Pay per Upload (PPU) — On the Free plan, opt any individual upload into PPU for that one file. Paid plans don’t expose PPU since storage is included.

See Payments for the full comparison.

OpenTusk is built around a single owner account that controls access for many agents. You set up once, then connect as many AI agents as you need — each with its own identity and scoped permissions.

You (owner)
├── Create account at app.opentusk.ai or via CLI
├── Create vaults (shared or public)
├── Generate invite codes for each agent
├── Agent 1 (Claude Code)
│ └── opentusk login --invite-code otinv_...
│ └── Gets its own API key + Sui keypair
├── Agent 2 (Cursor)
│ └── opentusk login --invite-code otinv_...
│ └── Gets its own API key + Sui keypair
└── Agent 3 (OpenClaw / CI pipeline / custom bot)
└── opentusk login --invite-code otinv_...
└── Gets its own API key + Sui keypair

Each agent gets a unique API key and Sui keypair, scoped to your account. You control what each agent can access — specific vaults, specific scopes, with automatic expiry.

  • Upload and download files — persist conversation context, generated artifacts, reports, code, or any binary data
  • Organize with vaults and folders — each agent can work in its own vault or share vaults with other agents
  • Read and write inline content — sandboxed agents (Claude Desktop) that can’t touch the filesystem can still create and read files through inline content tools
  • Collaborate through shared vaults — multiple agents and humans can read and write to the same encrypted vault, with access controlled by Sui addresses on-chain
  • Manage trash and lifecycle — soft-delete, restore, and track file sync status across hot cache and Walrus
  • Vault creation and visibility — only the owner can create and delete vaults
  • Access grants — the owner decides which Sui addresses (agents or humans) can access each shared vault
  • API key and invite management — create, scope, and revoke agent credentials at any time
  • Billing and quotas — all storage usage rolls up to the owner’s plan

The Install & Setup guide walks through this entire flow — from account creation to your first agent upload.