Use AgentGid from your AI agent

Add AgentGid as an MCP server and your agent can pick tools with live data: which API or agent fits a goal, what it costs and what's free, plan limits, whether something is down right now, and what developers actually use. Free, read-only, no sign-in.

Server URL (Streamable HTTP)https://agentgid.com/mcp

Add it to your agent

Claude Code

claude mcp add --transport http agentgid https://agentgid.com/mcp

Cursor

.cursor/mcp.json or Settings → MCP

{
  "mcpServers": {
    "agentgid": { "url": "https://agentgid.com/mcp" }
  }
}

VS Code (Copilot)

.vscode/mcp.json

{
  "servers": {
    "agentgid": { "type": "http", "url": "https://agentgid.com/mcp" }
  }
}

Codex

~/.codex/config.toml

[mcp_servers.agentgid]
url = "https://agentgid.com/mcp"

Claude.ai and ChatGPT

Add a custom connector with the server URL above (Settings → Connectors). No authentication.

Any other client

Clients that only speak stdio can use npx mcp-remote https://agentgid.com/mcp.

What your agent can ask

`find_tools`	Candidates for a goal: "copy-trading bot for Solana LP pools", "3D assets for a game", "web search for a research agent under $50 a month".
`find_apis`	Compare APIs in a category on official price, free tier, MCP support, SDK downloads and status; filter by price, chain or free tier.
`get_api` · `get_agent`	Full pricing, limits, packages, MCP server, status and cheaper alternatives.
`check_status`	Is Claude Code, Cursor, Helius or Tavily down right now? From official status pages.
`model_prices`	LLM prices per 1M tokens and which agents can use each model.
`real_world_usage`	MCP servers developers actually connect, from public GitHub repos.
`list_api_categories`	The 99 APIs we compare, by category.

Answers include links to the AgentGid pages behind them and the date prices were checked. Limit: 60 calls a minute per IP. The data is also on the site, in llms.txt and as JSON.