Free-first Clervo tools and bounded opt-in commerce through one local wallet.
Clervo lets software use AI models and agent tools with pay-per-use x402 payments, without managing separate provider accounts or API keys.
The public API is https://api.clervo.dev. Current products include AI, Web Search, Secure Sandbox, Prediction Intelligence, and Crypto Intelligence. Multi-chain RPC is not currently available through the public API.
Make a free AI call:
curl -sS https://api.clervo.dev/v1/chat/completions \
-H 'content-type: application/json' \
-d '{"model":"clervo/laguna-s-2.1","messages":[{"role":"user","content":"Reply with ready."}],"max_completion_tokens":16}'
Connect Claude Code through MCP:
claude mcp add clervo -s user -- npx -y @clervo/mcp
Or choose another supported path at clervo.dev/start:
@clervo/sdk for TypeScriptclervo-sdk for Python@clervo/mcp for Claude and other MCP clients@clervo/router for the CLI and local OpenAI-compatible proxyThe hosted API supports OpenAI Chat Completions at POST /v1/chat/completions, OpenAI Responses at POST /v1/responses, Anthropic Messages at POST /v1/messages, and the native Clervo route at POST /v1/ai/execute.
Free operations require no wallet. Paid operations return HTTP 402 with the maximum price before execution. Clervo clients keep automatic payment off by default and support per-operation limits, daily limits, receipts, reconciliation, and same-key replay without a second payment.
Payment uses USDC on Base through x402 or MPP. If settlement is unknown, retry only after reconciliation; never create a new authorization just to recover an uncertain request.
Current products, model availability, routes, and prices are machine-readable:
The repository requires Node.js 24.18.1 and npm 10.9.8.
npm ci --ignore-scripts
npm run generate:discovery
npm run test:b13:clients
npm run site:build
Never place credentials, wallet material, customer payloads, or authentication files in issues or commits. Report security concerns through the repository's private vulnerability-reporting channel.
This source is currently unlicensed (UNLICENSED). No permission to copy, modify, or redistribute is granted unless Clervo publishes separate terms.
Source-derived launch command. Check the maintainer’s required arguments and credentials before running:
npx -y @clervo/mcpMerge this template into ~/Library/Application Support/Claude/claude_desktop_config.json. Keep existing servers. Add any arguments, credentials, and permissions required by the maintainer; this template has not been install-tested.
{
"mcpServers": {
"io-github-clervo-connect": {
"command": "npx",
"args": [
"-y",
"@clervo/mcp"
]
}
}
}Restart Claude Desktop completely for changes to take effect. Confirm the server appears connected in the client’s tool list, then try a read-only example from its documentation.
Claude Desktop setup reference@clervo/mcpnpmClervo Connect works with any MCP-compatible client. Copy the config snippet from the Configuration section above and add it to the file shown for your client, then restart the application.
~/Library/Application Support/Claude/claude_desktop_config.jsonRestart Claude Desktop completely for changes to take effect.~/.cursor/mcp.jsonRestart Cursor for changes to take effect..vscode/mcp.jsonReload VS Code window for changes to take effect.~/.codeium/windsurf/mcp_config.jsonRestart Windsurf for changes to take effect..mcp.jsonSave at the project root, then start Claude Code in that project and review the MCP server approval prompt. Keep real credentials out of shared files.