io.github.bridgenode-ai/bridgenode-mcp (BridgeNode MCP)

BridgeNode provides “x402 pay-per-request AI inference” via an OpenAI-compatible API and a Model Context Protocol (MCP) server. The service is described as Solana USDC-based, “gas-free,” with no API key mentioned in the topics. The package is published as @bridgenode/mcp under MIT-0.

🛠️ Key Features

OpenAI-compatible API access
MCP server integration
Pay-per-request (x402) inference model
Solana USDC-based usage; gas-free execution
Tools available: 3

🚀 Use Cases

LLM access through OpenAI-compatible clients
MCP-based agentic/autonomous workflows
Vision and agentic-commerce style use cases

⚡ Developer Benefits

MCP support for model-context tooling
Compatibility with OpenAI-compatible interfaces
Payment method tagged as pay-per-request
Topics include “no-api-key”

⚠️ Limitations

Source data does not specify tool names, authentication details beyond “no-api-key,” or supported model list.

{ "type": "object", "properties": { "model": { "type": "string", "description": "Explicit model id (see list_models / /v1/models). Mutually exclusive with mode." }, "mode": { "type": "string", "enum": [ "auto", "eco", "premium" ], "description": "Smart routing profile (auto/eco/premium)." }, "messages": { "type": "array", "description": "Chat messages (text content only). An assistant turn that calls a tool may carry content: null with tool_calls.", "items": { "type": "object", "properties": { "role": { "type": "string", "enum": [ "system", "user", "assistant", "tool" ] }, "content": { "type": [ "string", "null" ] }, "tool_calls": { "type": "array" }, "tool_call_id": { "type": "string" } }, "required": [ "role" ] } }, "max_tokens": { "type": "integer", "minimum": 1, "description": "Max output tokens (billed upfront, §4.2). MCP calls are non-stream: cap per config. Use >= 200 — reasoning models share this budget with their thinking and a smaller limit can return an empty answer." }, "tools": { "type": "array", "description": "OpenAI-style tool schemas the model may call. Forwarded unchanged; they count as input tokens for the price (§4.2/§5.3).", "items": { "type": "object", "properties": { "type": { "type": "string", "enum": [ "function" ] }, "function": { "type": "object" } }, "required": [ "type", "function" ] } }, "tool_choice": { "description": "OpenAI tool_choice: 'auto', 'none', 'required', or {type: function, function: {name}}.", "type": [ "string", "object" ] } }, "required": [ "messages" ], "$schema": "https://json-schema.org/draft/2020-12/schema", "anyOf": [ { "required": [ "model" ] }, { "required": [ "mode" ] } ] }

@bridgenode/mcp

Built for AI agents — no API keys, no registration, pay-as-you-go with Solana USDC via x402. Start free: free models and 2 free calls on paid models per client, no wallet required.

BridgeNode MCP wrapper — stdio MCP server that proxies requests to the remote https://bridgenode.cc/mcp (streamable-HTTP) with automatic x402 payment (Solana USDC). No API keys, no registration.

Configuration (.env)

bash

# Solana wallet private key (base58) — agent's payment wallet
# Optional — Solana wallet private key (base58). Needed only for PAID calls:
# free models and the 2 free trials work without it.
BRIDGENODE_WALLET_KEY=

# Spending policy (fail-closed)
BRIDGENODE_MAX_PER_CALL=0.05
BRIDGENODE_DAILY_CAP=1.0

Tools

chat_completions — AI inference (free models run without payment; the first 2 calls to a PAID model are free per client, then x402 payment is automatic). Accepts OpenAI-style tools / tool_choice — the model's answer then carries tool_calls (the wrapper passes arguments through unchanged)

list_models — model list (free)

get_price_estimate — estimate cost of a chat request (free)

How it works

Local MCP client launches the wrapper via stdio

Wrapper connects to https://bridgenode.cc/mcp (streamable HTTP)

On tools/call receiving 402 (payment required) — the wrapper automatically signs the payment with BRIDGENODE_WALLET_KEY and retries the request

Spending policy checks every payment BEFORE signing (fail-closed)

Python packages

Prefer Python? The same BridgeNode toolkit is on PyPI:

Install

Tools (3)

chat_completionslist_modelsget_price_estimate

Add to VS Code Add to Cursor

claude_desktop_config.json

json

{
  "mcpServers": {
    "bridgenode-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@bridgenode/mcp@1.0.14"
      ]
    }
  }
}

BridgeNode MCP

io.github.bridgenode-ai/bridgenode-mcp (BridgeNode MCP)

🛠️ Key Features

🚀 Use Cases

⚡ Developer Benefits

⚠️ Limitations

Topics

chat_completions

list_models

get_price_estimate

@bridgenode/mcp

Installation (one-command)

Configuration (.env)

Tools

How it works

Links

Python packages

Community

Install

Related servers

model	string	optional	Explicit model id (see list_models / /v1/models). Mutually exclusive with mode.
mode	string	optional	Smart routing profile (auto/eco/premium).
messages	array	required	Chat messages (text content only). An assistant turn that calls a tool may carry content: null with tool_calls.
max_tokens	integer	optional	Max output tokens (billed upfront, §4.2). MCP calls are non-stream: cap per config. Use >= 200 — reasoning models share this budget with their thinking and a smaller limit can return an empty answer.
tools	array	optional	OpenAI-style tool schemas the model may call. Forwarded unchanged; they count as input tokens for the price (§4.2/§5.3).
tool_choice	string \| object	optional	OpenAI tool_choice: 'auto', 'none', 'required', or {type: function, function: {name}}.