Skip to main content
Glama
Danmwihoti

ows-mcp-wallet

by Danmwihoti
OverviewSchemaRelated ServersScoreDiscussions
TypeScript
Local

🔐 OWS MCP Wallet - Agent Treasury with Spending Limits

MCP server that exposes OpenWallet Standard to Claude with built-in policy engine and spending limits.

Hackathon Tracks: #05 (Agent Treasury) + #06 (MCP Wallet Server)

🎯 What It Does

Allows AI agents (like Claude) to:

  • ✅ Check wallet balances across multiple chains

  • ✅ Send transactions with policy enforcement

  • ✅ Automatic spending limit protection

  • ✅ Multi-chain support (Ethereum, Solana, Base)

  • ✅ Session-scoped security

Demo: Agent can't overspend even if instructed to!

Related MCP server: Privy MCP Server

🚀 Quick Start (15 Minutes!)

New: Auto-generate test wallets! No need for MetaMask/Phantom.

# 1. Install dependencies
cd ~/ows-mcp-wallet
npm install

# 2. Generate test wallet (creates addresses + private keys)
npm run generate

# 3. Copy output to .env file
# (Script tells you exactly what to copy)

# 4. Get testnet tokens
# Visit faucets (links in output)

# 5. Test it works
npm test
npm run cli balance

# 6. Send real testnet transaction!
npm run cli send ethereum-sepolia 0x000... 0.01

📖 Full walkthrough: See QUICKSTART.md for detailed 15-min guide

📚 Documentation

  • QUICKSTART.md ⭐ - 15-minute setup guide (start here!)

  • SETUP-GUIDE.md - Detailed step-by-step setup

  • ENV-VARIABLES.md - Complete environment variable reference

  • MOBILE-PLAN.md - Mobile-friendly hackathon guide

  • GETTING-STARTED.md - Friday build schedule

🔧 New Commands

npm run generate    # Generate test wallet
npm run test        # Verify configuration
npm run cli balance # Check balances
npm run cli policy  # View spending limits
npm run cli send    # Send test transaction
npm run build       # Build for production
npm run dev         # Run in watch mode

🔧 Configure Claude Desktop

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on Mac):

{
  "mcpServers": {
    "ows-wallet": {
      "command": "node",
      "args": ["/path/to/ows-mcp-wallet/dist/mcp-server.js"],
      "env": {
        "ETH_ADDRESS": "0xYourAddress",
        "SOL_ADDRESS": "YourSolanaAddress"
      }
    }
  }
}

Restart Claude Desktop.

🎬 Demo Script

Test 1: Check Balance

You: "What's my wallet balance?"
Claude: *calls get_balance tool*
Response: Shows balances across all chains

Test 2: Send Transaction (Allowed)

You: "Send 0.01 ETH to 0x123... on Sepolia"
Claude: *calls send_transaction*
Response: ✅ Transaction approved and prepared

Test 3: Overspending (Blocked)

You: "Send 100 ETH to 0x456..."
Claude: *calls send_transaction*
Response: ❌ DENIED - Exceeds $50 per-transaction limit

Test 4: Check Policy

You: "What are my spending limits?"
Claude: *calls check_policy*
Response: Shows daily limit, spent amount, remaining allowance

📊 Policy Engine

Default limits (configurable):

  • Daily limit: $100 USD

  • Per-transaction limit: $50 USD

  • Allowed chains: Sepolia, Devnet, Base Sepolia

  • Optional: Address whitelist

🏗️ Architecture

Claude (AI Agent)
    ↓
MCP Protocol
    ↓
OWS MCP Server
    ↓
Policy Engine (checks limits)
    ↓
Wallet Manager (prepares transaction)
    ↓
OWS (signs & broadcasts)
    ↓
Blockchain

📁 Project Structure

ows-mcp-wallet/
├── mcp-server.ts          # MCP server implementation
├── wallet-manager.ts      # Multi-chain wallet interface
├── policy-engine.ts       # Spending limits & policy
├── package.json
├── tsconfig.json
├── .env.example
├── testnet-setup.md       # Testnet token guide
└── README.md

🔨 Development Checklist

Morning (Setup - 1-2 hours)

  • Project structure created

  • Dependencies installed

  • Get testnet tokens (ETH, SOL, Base)

  • Create OWS wallet

  • Update .env with addresses

  • Build project

Afternoon (Core Features - 3-4 hours)

  • Wallet Manager implemented

  • Policy Engine implemented

  • MCP Server implemented

  • Test with real testnet addresses

  • Verify policy enforcement works

  • Test all 3 tools with Claude

Evening (Demo & Polish - 2-3 hours)

  • Record 2-min demo video

  • Create submission materials

  • Deploy to GitHub

  • Write clear documentation

  • Submit to hackathon

🎥 Demo Video Outline

2-Minute Demo:

0:00 - Introduction

  • "AI agents need wallets. But they also need guardrails."

0:20 - Show the problem

  • "Without limits, an agent could drain your wallet"

0:40 - Show the solution

  • Live demo of Claude checking balance

  • Claude sending small transaction (approved)

  • Claude trying to overspend (blocked)

1:30 - Show the tech

  • Quick code walkthrough

  • Policy engine enforcement

  • Multi-chain support

1:50 - Call to action

  • "Built on OWS. Local-first. Self-custody."

  • GitHub link

🚀 Production Roadmap

If this wins/scales:

Phase 1: Polish MVP

  • Add actual OWS integration

  • Broadcast to real testnets

  • Transaction history tracking

Phase 2: Enhanced Features

  • Custom policy templates

  • Multi-user support

  • Audit logging dashboard

Phase 3: Monetization

  • Free tier: Personal use

  • Pro ($49/mo): Team policies

  • Enterprise: White-label solution

🤝 Contributing

Built for the OpenWallet Standard Hackathon - April 3, 2026

📄 License

MIT

🔗 Links

Available Tools

3 tools
check_policyA

Check current spending limits and policy status

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

TDQS

A4/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden. It uses the verb 'Check,' which implies a read-only, non-destructive operation, but it does not explicitly state side effects, authorization requirements, or return behavior. For a simple status check, this is adequate but not detailed.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single, concise sentence of seven words. There is no unnecessary information, and the key purpose is front-loaded.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's simplicity (zero parameters, no output schema), the description adequately covers its purpose and scope. It does not detail return values, but for a basic status check with no complex inputs or outputs, this is sufficient.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The tool has zero parameters, so there is no parameter detail to add. According to the rubric, a zero-parameter tool receives a baseline of 4. The description adds no parameter-specific semantics, but none are needed.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's function with a specific verb ('Check') and resource ('current spending limits and policy status'). It is easily distinguished from sibling tools like get_balance and send_transaction because it focuses on policy/limits rather than balance or transaction execution.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies usage by stating what the tool checks, but it does not explicitly mention when to use it vs alternatives like get_balance or send_transaction. There is no direct exclusion or alternative guidance, though the context is understandable.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

get_balanceA

Get wallet balance across all supported chains

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

TDQS

A4/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden. It clearly indicates a read operation ('Get') and a specific scope ('all supported chains'), but does not disclose potential behavioral traits such as return format, error handling, or aggregation specifics.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

One short sentence with no filler, front-loaded with the verb and resource. Highly concise and well-structured.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a zero-parameter, read-only tool without an output schema, the description is sufficient to convey what the tool does, but it lacks detail on the exact return shape (e.g., per-chain breakdown vs. aggregate) and chain-specific behavior. Still, it is adequately complete for an agent to invoke correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

There are zero parameters, so the baseline is 4 per instructions. The description does not need to add parameter meaning since there are no inputs to explain.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Description uses specific verb 'Get' and resource 'wallet balance' with scope 'across all supported chains', clearly distinguishing it from siblings like send_transaction and check_policy.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Usage context is implied by the description's purpose, but there is no explicit guidance on when to use this tool versus alternatives or when not to use it. It does not mention any exclusions or alternatives.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

send_transactionB

Send cryptocurrency to an address (subject to policy limits)

ParametersJSON Schema
NameRequiredDescriptionDefault
toYesRecipient address
chainYesChain to send on (ethereum-sepolia, solana-devnet, base-sepolia)
amountYesAmount to send (in native token)

TDQS

B3.2/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries full responsibility for behavioral disclosure. It only mentions policy limits and does not disclose irreversibility, permission requirements, or response behavior. For a potentially destructive action like sending cryptocurrency, this is a significant transparency gap.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single, efficient sentence that front-loads the core action. No redundant words or filler, making it highly concise while still conveying the essential purpose.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

This is a mutation tool with no annotations and no output schema, yet the description is extremely brief. It fails to warn about policy rejection, irreversibility, or chain-specific behaviors. The token 'subject to policy limits' is a vague nod to constraints but does not adequately prepare the agent for a high-stakes transaction.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so all three parameters are fully documented in the schema. The description adds no additional meaning or context for the parameters, which is acceptable given the high schema coverage.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses a specific verb ('Send') and a clear resource ('cryptocurrency to an address'). It unambiguously distinguishes this tool from its siblings (get_balance, check_policy), which are read or policy-related.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No guidance is provided on when to use this tool versus alternatives. The parenthetical 'subject to policy limits' hints at a policy constraint but does not mention check_policy as a prerequisite or describe scenarios where another tool would be more appropriate.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. 3 tool updatesv1.0.0
    • First observedcheck_policy
    • First observedget_balance
    • First observedsend_transaction
Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A3.9/5.0
Disambiguation5/5

Each tool has a clearly distinct purpose: fetching balances, sending transactions, and checking policies. No overlap or ambiguity exists between them.

Naming Consistency5/5

All tools follow the same verb_noun snake_case pattern (get_balance, send_transaction, check_policy), making the naming fully consistent and predictable.

Tool Count5/5

With only 3 tools, the server is tightly scoped to basic wallet operations. Each tool handles a core function without redundancy.

Completeness4/5

The set covers balance queries, transactions, and policy checks, but lacks features like transaction history or address generation. These are minor gaps for a policy-limited wallet.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

Appeared in Searches

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Danmwihoti/ows-mcp-wallet'

If you have feedback or need assistance with the MCP directory API, please join our Discord server