Install
openclaw skills install @bridgenode/bridgenodeBridgeNode — anonymous pay-per-request LLM inference for AI agents without API keys or registration. OpenAI-compatible chat completions endpoint + MCP access...
openclaw skills install @bridgenode/bridgenodeBridgeNode is an AI inference service for agents: anonymous LLM access without API keys, without registration, without subscriptions. Agents get an OpenAI-compatible chat completions endpoint and pay as they go with Solana USDC micropayments via the x402 payment protocol (HTTP 402). Model prices are published per token on the public models endpoint; transaction fees are sponsored, so an agent only needs USDC in its own wallet. Works with any OpenAI-compatible agent, MCP clients, and x402-capable SDKs.
This skill spends real money (USDC) on every request via on-chain Solana USDC micropayments (x402). There are no refunds once the provider has responded. Check live prices at GET /v1/models before use, set max_tokens to control cost, and use client-side spending limits (BRIDGENODE_MAX_PER_CALL, BRIDGENODE_DAILY_CAP) if available.
BridgeNode is an AI inference bridge. Agents get an OpenAI-compatible endpoint and pay per request with Solana USDC via the x402 protocol. No API keys, no registration, no subscriptions. Transaction fees (SOL) are sponsored by BridgeNode — the agent only needs USDC in its wallet.
| Endpoint | Purpose |
|---|---|
https://bridgenode.cc/v1 | OpenAI-compatible API base URL |
https://bridgenode.cc/v1/models | Public model list + prices (no auth) |
https://bridgenode.cc/v1/chat/completions | Chat completions (POST) |
https://bridgenode.cc/mcp | MCP server (streamable-http) |
https://bridgenode.cc/llms.txt | Full agent install map |
Prices are in USDC per token (6 decimals). Always fetch live prices from GET /v1/models — they are the single source of truth and are generated from server config (never hardcoded here — stale prices cost money).
🆓 Free models (no payment, no API key, no registration): gpt-oss-20b · gpt-oss-120b · glm-4.7-flash · glm-4.5-flash · glm-4.6v-flash (vision).
Paid models (pay-per-request): DeepSeek, GLM (Z.AI), Kimi (Moonshot), MiniMax. Full list with live prices: GET https://bridgenode.cc/v1/models.
Pricing model: exact scheme — the agent pays for input tokens + max_tokens before processing. If the model generates fewer than max_tokens, the agent still pays for max_tokens (this is the business model, not a bug). Minimum charge per request: 2000 atomic units = $0.002 USDC.
max_tokens budget with the answer.max_tokens >= 200 — a too-small limit can be fully consumed by reasoning, producing an EMPTY answer (the model returned 200 with no content).max_tokens and purchase again.stream: true for long generations (non-stream is capped at 4096).reasoning_content in the next turn, otherwise the API returns 400.402 Payment Required with a PAYMENT-REQUIRED header (base64 JSON): price, payTo address, USDC mint, memo, recent blockhash.TransferChecked (amount = required) + Memo instruction, signs with its own wallet. Fee payer is NOT signed by the agent.PAYMENT-SIGNATURE header (base64 JSON payload with the signed transaction).200 with PAYMENT-RESPONSE header (settlement receipt).Key details:
solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp (Solana mainnet)EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v@x402/svm, x402[svm]) or any x402-capable client — they handle the 402 → sign → retry flow automatically.BRIDGENODE_MAX_PER_CALL, BRIDGENODE_DAILY_CAPStep 1 — get payment requirements:
curl https://bridgenode.cc/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{"model":"deepseek-v4-flash","messages":[{"role":"user","content":"hello"}],"max_tokens":100}'
Response: 402 with PAYMENT-REQUIRED header (amount, payTo, memo).
Step 2 — sign the partial transaction with an x402-capable client (e.g. x402-proxy (npx x402-proxy), official SDK, or pay CLI) and retry:
curl https://bridgenode.cc/v1/chat/completions \
-H "Content-Type: application/json" \
-H "PAYMENT-SIGNATURE: <base64 payload>" \
-d '{"model":"deepseek-v4-flash","messages":[{"role":"user","content":"hello"}],"max_tokens":100}'
Response: 200 with the completion and PAYMENT-RESPONSE header.
# pip install "x402[svm]"
# Official x402 client handles the full 402 -> sign -> retry loop.
Use the official x402 Python client (x402[svm]) configured with the agent's Solana keypair; point it at https://bridgenode.cc/v1/chat/completions. Payment is automatic.
pip install bridgenode-llm (https://pypi.org/project/bridgenode-llm) — full toolkit: pip install bridgenodepip install bridgenode-cli (https://pypi.org/project/bridgenode-cli) — bridgenode chat "Hello!"npm i @bridgenode/llm (https://www.npmjs.com/package/@bridgenode/llm)npm i @bridgenode/mcp (https://www.npmjs.com/package/@bridgenode/mcp)All SDKs handle the x402 payment handshake automatically (402 → sign → retry → 200), with fail-closed spending limits (BRIDGENODE_MAX_PER_CALL, BRIDGENODE_DAILY_CAP).
claude mcp add bridgenode -s user -- npx -y @bridgenode/mcp@latesthttps://bridgenode.cc/mcp (streamable-http)chat_completions (model, mode, messages, max_tokens)tools/list (x-x402) as an indication — always check the actual amount in the 402 response before signing.model: explicit model ID from /v1/models (e.g. deepseek-v4-flash).mode: smart routing — auto (complexity-based tier), eco (cheapest), premium (best). If both model and mode are sent, model wins.max_tokens: request cap (default 4096, clamped to model max). Non-stream requests are capped at 4096 — use stream: true for longer generations.stream: SSE streaming supported (stream: true).| Status | Meaning |
|---|---|
| 400 | Bad request (unknown model, invalid body, oversized non-stream max_tokens) |
| 402 | Payment required — see PAYMENT-REQUIRED header |
| 413 | Request body too large (limit 2 MB) |
| 429 | Too many requests (queue limit) |
| 503 | Service busy — retry with backoff |
All errors use the OpenAI error format: {"error": {"message": ..., "type": ..., "code": ...}}.
Security: ClawHub security audit Pass; VirusTotal scan clean (no engine findings).
Discovery: https://bridgenode.cc/.well-known/agent-card.json, /.well-known/mcp.json, /.well-known/ai-manifest.json
Listed on x402-list: https://x402-list.com/services/bridgenode
Listed on x402-dev: https://www.x402dev.com/awesome-projects/
Listed on nohumans.directory: https://nohumans.directory/l/f1f74751-9d5
Listed on gold-402: https://github.com/Haustorium12/gold-402/blob/main/directory/learning.md
ClawHub skill: https://clawhub.ai/bridgenode/skills/bridgenode
Transaction fees are sponsored (gasless) — the agent only needs USDC in its own wallet.
Refunds: if the provider fails before any content is delivered, the payment is refunded automatically (reverse USDC transfer).