mcpField
C
Cloud Wallet by Ekubo

Cloud Wallet by Ekubo

Remote
cloud.ekubo.orgFinance0Updated 6d ago

About Cloud Wallet by Ekubo

Cloud Wallet by Ekubo is a hosted EVM wallet for AI agents to inspect accounts, simulate transactions and request execution. Owner-defined policies and approval checks govern transaction requests.

Key features

  1. 01Discover accounts and policies
  2. 02Read token balances and allowances
  3. 03Read contract data
  4. 04Simulate execution plans
  5. 05Request transaction execution
  6. 06Track execution status

Use cases

  1. 01Inspect accessible wallet accounts
  2. 02Check token balances before acting
  3. 03Simulate a producer's execution plan
  4. 04Monitor a transaction request

MCP support

It provides a hosted MCP server over remote Streamable HTTP with OAuth authorization. MCP clients can access wallet tools according to owner-granted scopes and account policies.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://cloud.ekubo.org/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

Cloud Wallet by Ekubo

Cloud Wallet is a hosted EVM wallet that lets AI agents read account data, simulate transactions, and request transaction execution under rules chosen by the owner. This repository contains setup and usage documentation for the hosted MCP server.

Connect

  1. Open Cloud Wallet to set up your wallet and account policies.
  2. Add https://cloud.ekubo.org/mcp as a remote Streamable HTTP MCP server in your client.
  3. Follow the client's OAuth authorization link. The owner signs in, chooses the accounts and access to grant, and consents. Choose read access for account queries or write access when execution is needed. Write access includes read access at the same scope.
  4. Ask the agent to list the accounts available to the connection. It should call wallet_list and use the returned account identifiers for subsequent account requests.

Your client must support remote MCP connections and the server's OAuth flow. Producer MCP servers are separate connections; wallet_list_integrations lists supported producers and their connection URLs.

Capabilities

TaskTools
Discover accessible accounts and inspect their policieswallet_list, wallet_get_account
Read token balances and allowanceswallet_fetch_balances
Read contractswallet_batch_eth_call
Simulate a producer's execution planwallet_simulate_execution_plan
Request execution of a successful simulationwallet_send_execution_plan
Track a request through its final statuswallet_get_execution_status

Cloud Wallet has financial transaction capabilities. Requests are evaluated against account policies and service checks. Depending on those results, a request can execute automatically, be rejected, or wait for the owner's passkey approval.

Example requests

  • “List the accounts this connection can access.”
  • “Show the policy for this account.”
  • “Read this account's token balances.”

For an authorized transaction, connect the relevant producer MCP server, obtain its execution plan, and pass that plan unchanged to Cloud Wallet for simulation. A successful, sendable simulation provides a one-use simulation_id valid for 120 seconds. Send that ID with a fresh UUID idempotency key, then check request status until it is confirmed, failed, rejected, or cancelled. A pending_review result waits for the owner in the Cloud Wallet dashboard.

Use the server's live tool descriptions and instructions for the current arguments, permissions, limits, and recovery steps.

Documentation

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.