Introduction

Monid lets your AI agent find and use agentic tools on demand. Your agent discovers the right tool, checks pricing, and executes it. You pay only for what you use.

How It Works

1. Discover what tools are available for a given task
2. Inspect a tool's input schema, pricing, and documentation
3. Run the tool with structured input and retrieve results

Monid handles trust, payment, and fulfillment between your agent and agentic tools.

Supported Endpoints and Tools

The catalog grows every single day. Use monid discover or POST /v1/discover to find what endpoints and tools are available for your given task.

How to Use Monid

MCP

Connect Claude.ai, ChatGPT, or any MCP-compatible AI client via Streamable HTTP. No installation needed — just add the server URL and log in.

https://mcp.monid.ai/v1

→ More detail in quick setup

Skill

Paste the following instruction to your AI agent (Claude Code, Cursor, etc.) to setup Monid skill. Your agent learns to discover, inspect, and run tools automatically.

set up https://monid.ai/SKILL.md

→ More detail in quick setup

CLI

Install the Monid CLI for terminal use. Works for both human developers and AI agents.

npm install -g @monid-ai/cli

→ Quick setup · CLI Reference

HTTP API

Call the Monid API directly from code or agent frameworks.

curl -X POST https://api.monid.ai/v1/discover \
  -H "Authorization: Bearer <API Key>" \
  -H "Content-Type: application/json" \
  -d '{"query": "twitter posts", "limit": 5}'

→ API Reference

OAuth

Building a platform that calls Monid on behalf of your users? Connect via OAuth 2.0 and call the authenticated API with a user's access token.

→ OAuth setup

Next Steps

• Pricing -- How endpoints are priced