Veyline MCP Server
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Veyline MCP Server preflight this x402 quote before my wallet signs"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Veyline by Payload — MCP Server
An MCP server (stdio transport, Node.js standard library only, zero dependencies) that exposes Veyline, Payload's flagship noncustodial authorization layer, to AI agents.
Veyline decides whether an agent payment may proceed, under policy, before any wallet signs. This server wraps the Payload rail's Veyline A2A skills as MCP tools. Every tool call goes to the live Payload rail (https://payload-rail.fly.dev) authenticated with your own rail API key, for your own existing Veyline organization.
Tools
Tool | Upstream route | What it does |
|
| Read the current authenticated organization's Veyline entitlement and tier. Does not authorize a payment. |
|
| Read the current authenticated organization's quota and usage. Does not change policy. |
|
| Request a signed, short-lived, policy-enforced payment preflight. No funds are transferred. Requires Veyline execution scope and the original authoritative x402 quote. Inputs include the authoritative |
Related MCP server: OnChainDiligence Agentverse Adapter
Adding Veyline to an agent workflow
Veyline sits between an agent that wants to pay and the wallet that signs. You add it to the stack you already have; you do not rebuild around it.
The agent identifies a payment it intends to make. Your application knows the recipient, amount, asset, and network.
The application obtains authoritative payment details. For x402-priced actions, that is the original quote returned by the seller, not a number the agent guessed.
Veyline evaluates the proposed action. Call
veyline_preflightwith the complete authoritative quote and its associated request hash, action, amount in cents, recipient, asset, CAIP-2 network, and quote ID. Incomplete quote requests fail closed. Veyline checks the request against your organization's authorization policies, spending limits, and permitted recipients before anything is signed.An eligible request receives a scoped, signed, expiring authorization. A preflight returns a signed, short-lived authorization bound to that payment intent. A request outside policy is refused, and nothing proceeds.
Your own wallet or signer executes the payment. The customer's wallet, signer, or approved execution system performs the payment within the authorized boundaries. Veyline never holds funds and never signs in place of your wallet.
Evidence is recorded or verified through supported functionality. Your application keeps the preflight result alongside the payment it authorized. On the same rail, RevRule records payout evidence linked back to the original entitlement entries, so a payout can later be checked against what was owed.
Two reads round out the integration: veyline_status reports your organization's entitlement and tier, and veyline_usage reports quota and usage. Neither authorizes a payment.
Plainly stated: Veyline provides governance and authorization without custody of customer funds. It does not execute payments itself, and it cannot guarantee that external payment systems, networks, or counterparties will not fail. What it does guarantee is narrower and useful: a payment your application routes through preflight gets a policy check before any wallet signs.
Evaluate free, then subscribe
Evaluation: get a free Payload rail API key from the sandbox signup on https://payloadhq.github.io/veyline.html (email only, free tier with 1,000 operations per month) or from the console at https://payload-rail.fly.dev/revrule-console/, and use it with your Veyline organization. Install the verified MCPB release below; no preconfigured key is included.
Paid: the Veyline Developer subscription starts at $99 per month via https://payloadhq.github.io/veyline.html and is the production path for payment-enabled agents and APIs that need policy-enforced authorization in the loop.
Install
Verified MCPB release
Install the veyline v1.1.0 MCPB bundle in an MCPB-compatible client, or run from source below. The published bundle declares the supported MCP tools. This direct release is the verified installation path while third-party directory listings are revalidated.
From source
git clone https://github.com/Payloadhq/veyline-mcp
cd veyline-mcp
node mcp/index.jsOr run it straight from GitHub with npx:
npx -y github:Payloadhq/veyline-mcpMCP client configuration (stdio):
{
"mcpServers": {
"veyline": {
"command": "node",
"args": ["/path/to/veyline-mcp/mcp/index.js"],
"env": {
"PAYLOAD_RAIL_API_KEY": "your-rail-api-key"
}
}
}
}Configuration
Variable | Required | Default | Purpose |
| Required for tool calls | — | Your own Payload rail API key for an existing Veyline organization. Sent to the rail only as a Bearer |
| Optional |
| Payload rail base URL. |
Fail-closed: with no API key configured, the server still starts and lists its tools, and every tool call returns a clear API key not configured error. Nothing is ever sent anywhere without your key, and the key never leaves your machine except as the Authorization header to the rail.
Notes
A preflight is an authorization check, not a payment: it never moves funds.
You need an existing Veyline organization on the Payload rail for the tools to return data.
License
MIT — see LICENSE.
Veyline is the flagship product of Payload. See also callx402 by Payload, the MCP client layer for x402-priced actions.
Available Tools
3 toolsveyline_preflightA
Request a signed, short-lived, policy-enforced payment preflight. No funds are transferred. Requires Veyline execution scope and the original authoritative x402 quote. Pass the payment intent (recipient, amount, asset, network) and, when available, the original x402 quote object. (Rail A2A skill veyline_preflight.)
Inputs: asset (required), quote (required), action (required), network (required), +8 more.
veyline_statusB
Read the current authenticated organization Veyline entitlement and tier. Does not authorize a payment. (Rail A2A skill veyline_status.)
Inputs: no named parameters.
veyline_usageB
Read the current authenticated organization quota and usage. Does not change policy. (Rail A2A skill veyline_usage.)
Inputs: no named parameters.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 3 tools
The three tools are mostly distinct: status reads entitlement/tier, usage reads quota consumption, and preflight requests a signed payment authorization. Status vs. usage overlap somewhat (both are read-only org-state reads about the same domain), but the descriptions make the boundary clear enough to select correctly.
All tools follow the same predictable pattern: a shared veyline_ prefix with a single snake_case scope noun (status, usage, preflight). No mixed conventions or stylistic outliers.
Three tools is on the thin side but defensible for a narrow read-only entitlement/usage plus preflight surface. Each tool earns its place with no redundancy.
The surface covers reads (entitlement, usage) and preflight authorization, but there is no tool to check preflight status, revoke/cancel a preflight, or submit/settle the actual payment path, leaving a lifecycle dead end after preflight. Agents must complete payment execution elsewhere.
Maintenance
Related MCP Connectors
Read-only checks for proposed orders against evidence and policy; no trading or payment authority.
Advisory policy preflight for AI-agent spend requests; never executes payments or accesses wallets.
Contract-change authorization for AI agents: preflight a change to an API or MCP contract, get a signed decision, and verify the receipt offline. Only a granted change can proceed.
Read-only discovery for exact-commit Agent Skill validation, x402 payment, and signed receipts.
Related MCP Servers
- AlicenseAqualityDmaintenanceEnables agents to request payment authorization against signed human mandates, enforcing policy-as-code limits and returning a decision (authorize, challenge, or deny) without moving funds.11Apache 2.0
- FlicenseNot gradedqualityCmaintenanceProvides a stateless Agent Chat Protocol bridge that lets agents inspect proposed payments against policy, retrieve public signed receipts, and verify their cryptographic proofs without wallet or payment capabilities.-
- AlicenseNot gradedqualityFmaintenanceEnables MCP clients to ask permission before an agent acts, returning allow, deny, or ask_human with receipts, and to monitor agent spend, limits, and recent decisions.57 npmFunctional Source , Version 1.1, ALv2 Future
- AlicenseAqualityAmaintenanceEnables agents and developers to check x402 endpoints, assess scores and wallet risks, authorize payments with spending rules, and report outcomes before an agent pays.62MIT