$ loading_
Build paid MCP servers that charge AI agents per call in USDC.
The available material is sparse, but it describes a paid MCP server that charges per call and handles on-chain-style payments. Although no keys or remote endpoints are declared, its code-execution capability and payment context warrant caution around actual payment flows, data egress, and local execution boundaries. Open source with an MIT license is a positive signal, but low adoption and unknown maintenance keep the overall posture at caution.
The material states that no keys or environment variables are required, and no explicit API token configuration is shown. However, its stated function includes USDC charging and payment handling via HTTP 402 and EIP-3009, so real deployments may still involve wallets, signing, or other payment-sensitive state. The current material is insufficient to rule out implicit credential or authorization abuse risks.
The material declares no remote endpoints, but 'per-call charging' and 'automatic payment handling' commonly imply interaction with payment-related systems or chain/gateway infrastructure. The documentation does not disclose what it connects to, whether request content is sent out, or the scope of payment metadata disclosure. With no README or endpoint list, this remains a caution item.
The system flags this tool as executes-code, meaning it can run code or start a local service process. That is a normal capability for this class of tool and not a high-risk red flag by itself. The material does not specify what additional system capabilities it can invoke, so it should be run in a constrained environment and its actual permission boundaries should be verified.
The description does not specify a concrete data read/write scope, nor does it declare access to particular local files, databases, or external resources. However, as an MCP server handling billed requests, it may naturally touch call parameters, billing records, or payment state. There is no evidence here of data access beyond its stated purpose, but the documentation is too sparse, so actual file and storage access should be inspected before use.
Positive signals include an auditable open-source repository under the MIT license, which lowers risk. However, the source is a third-party registry, community adoption is 0 stars, maintenance status is unknown, and there is virtually no README or installation detail, leaving limited audit context. No explicit malicious indicators are present, but the supply-chain maturity and verifiability are weak, so dependencies and commit history should be reviewed carefully.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "onyx-paid-mcp" yet — see the docs or source repo.
Help me design a paid MCP server that wraps an internal data query API. Charge a fixed USDC fee per call and handle payment automatically via HTTP 402 and EIP-3009. Provide the service architecture, request flow, and key code examples.
A paid MCP server plan with billing flow, payment handling logic, and sample code.
I have an MCP tool for web scraping and want AI agents to automatically pay USDC on every call. Generate a minimal working implementation with HTTP 402 responses, EIP-3009 authorized payments, and a client usage example.
A practical minimal implementation showing how to add automatic billing to an MCP tool.
Create a deployment and risk-control plan for a USDC per-call paid MCP server, including payment failure handling, retry strategy, rate limiting, audit logging, and security considerations.
A production-oriented deployment and risk-control checklist for launching a paid MCP service.
Enable AI agents to manage wallets and send or receive USDC payments.
Lets AI agents provision, manage, and renew cloud servers with USDC.
Integrate crypto payment requests, webhook verification, and payment URL generation.
Build, run, and debug real 2D and 3D games via AI conversation.
Enable agents to charge and accept payments securely via MCP and REST.
Enable non-custodial USDC payments on Base for AI agents via MCP.