proxy-mcp_skill

This skill helps you manage Proxy MCP payments by creating intents, issuing cards, and tracking balances and transactions.
  • Python

2.5k

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill proxy-mcp

  • _meta.json271 B
  • SKILL.md2.6 KB

Overview

This skill integrates an AI agent with Proxy's Model Context Protocol (MCP) to manage payments and virtual cards. It enables creating payment intents, provisioning virtual cards, checking balances, and tracking transactions using a Proxy account and agent token. The skill supports local NPX server launch or direct HTTP transport to Proxy's MCP endpoint.

How this skill works

The skill communicates with Proxy MCP tools to query user and KYC status, fetch balances, create payment intents that trigger card issuance, and retrieve card and transaction details. Typical flow: check balance, create an intent, fetch sensitive card details for payment, then monitor transactions. It surfaces standard Proxy error codes so calling code can handle onboarding, permissions, and funding issues.

When to use it

  • You need programmatic payment capability for an AI agent (one-time or recurring purchases).
  • You want to provision single-use or virtual cards for agent-driven payments.
  • You must track spending and transactions per agent or card.
  • You need to validate KYC/onboarding and surface instructions for funding.
  • You want to run a local MCP server via NPX during development.

Best practices

  • Always call proxy.balance.get before creating intents to avoid INSUFFICIENT_BALANCE errors.
  • Use proxy.kyc.status and proxy.kyc.link to ensure onboarding is complete before payments.
  • Assign a policy with spending limits to each agent to prevent runaway costs.
  • Store agent tokens securely and prefer scoped tokens with least privilege.
  • Handle Proxy error codes explicitly and surface actionable messages to operators.

Example use cases

  • An agent purchases API credits or SaaS subscriptions using a virtual card provisioned at intent creation.
  • Create one-time virtual cards tied to a single purchase to reduce fraud surface.
  • Report and reconcile transactions per agent by listing transactions for a card.
  • Automate funding instructions and display ACH/wire/crypto details to treasury staff.
  • Run a local NPX-based MCP server during integration testing to avoid remote calls.

FAQ

Yes. A Proxy account, completed KYC, an agent with an assigned policy, and an agent token are required.

How do I run the MCP server locally?

Use the NPX option to launch proxy-mcp-server with PROXY_AGENT_TOKEN set in the environment, or configure the HTTP transport to Proxy's MCP URL with Authorization header.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational