Skip to main content

Quick start

Type or paste this into any AI agent:
The agent will fetch the onboarding page, understand what the server offers, and walk you through setup. That’s it.
The Chainstack MCP server gives AI agents direct access to the Chainstack platform — deploy blockchain nodes, search documentation, request testnet faucet tokens, fetch live pricing, contact the Chainstack team, and check platform status, all from your coding environment. It works with Claude Code, Cursor, Codex, Windsurf, Gemini CLI, GitHub Copilot, Antigravity, Claude.ai, and ChatGPT. MCP (Model Context Protocol) is an open standard that lets AI applications connect to external tools and data sources. The Chainstack MCP server is a remote Streamable HTTP server — no local setup required.

Tools

No API key needed: API key required (get one at console.chainstack.com/user/settings/api-keys). All tools appear in the agent’s tool list regardless of auth — the tools below execute only when the key is configured. Calls without it return an auth error.

Option 1: Install as a skill

Save the SKILL.md to your agent’s skills directory. The agent calls tools via HTTP on demand — no MCP registration, no tools forced into context.
You may need to restart your agent or start a new session for the skill to take effect. The skill contains workflows, tool reference, curl examples, and error handling. To set up your API key for authenticated tools, add it to your shell profile:

Option 2: Register as an MCP server

Add the server to your agent’s configuration. Tools appear in context every session.

Claude Code

Codex CLI

Or edit ~/.codex/config.toml:

Cursor

Edit ~/.cursor/mcp.json (global) or .cursor/mcp.json (project):

Windsurf

Edit ~/.codeium/windsurf/mcp_config.json:

GitHub Copilot (VS Code)

Edit .vscode/mcp.json (workspace) or use Command Palette > “MCP: Open User Configuration” (global):
GitHub Copilot uses servers as the root key, not mcpServers.

Gemini CLI

Or edit ~/.gemini/settings.json (global) or .gemini/settings.json (project):
Gemini CLI uses httpUrl for Streamable HTTP endpoints, not url.

Antigravity

Edit ~/.gemini/antigravity/mcp_config.json:

Claude.ai

Settings > Connectors > Add custom connector > paste https://mcp.chainstack.com/mcp

ChatGPT

Settings > Apps & Connectors > Developer mode > Create > paste https://mcp.chainstack.com/mcp

Discovery

Visit mcp.chainstack.com for an agent-readable onboarding page. Point any agent at this URL (e.g., “get mcp.chainstack.com”) and it can read the page and self-configure.

Endpoints

Report issues

Found a bug or something not working as expected? Open an issue: Include what tool you called, what you expected, and what happened instead.

Request features

Have an idea for a new tool or improvement? Open a feature request: Describe your use case — what you’re trying to accomplish and why the current tools don’t cover it.

Changelog and updates

The changelog is maintained in the repository: To get notified of new releases and changes, watch the repository:
  1. Go to chainstacklabs/mcp-server
  2. Click Watch > Custom > select Releases
  3. You’ll receive notifications when new versions are published
Last modified on May 6, 2026