Documentation Index

Fetch the complete documentation index at: https://docs.konbiniapi.com/llms.txt Use this file to discover all available pages before exploring further.

MCP overview

Use KonbiniAPI directly from AI agents and coding assistants via the Model Context Protocol. No API calls, no headers — just ask.

KonbiniAPI is available as a hosted MCP server at https://mcp.konbiniapi.com. That lets AI agents like Claude, ChatGPT, Cursor, Codex, and Zed call KonbiniAPI directly without you writing integration code or running the server locally.

Once connected, your AI agent can fetch TikTok and Instagram data just by being asked. No manual API calls, no auth headers, no JSON parsing.

What the agent can do

The MCP server exposes one typed tool per supported endpoint:

Use tools like `instagram.get_user` or `tiktok.get_user_videos` directly in your client. Use `projection_preset`, `data_fields`, and `item_fields` to keep tool results focused and save tokens.

How it works

Connect your MCP-compatible client to https://mcp.konbiniapi.com, authorize it once, and then call KonbiniAPI tools directly from your agent. Most clients support OAuth. Some clients also let you connect with a KonbiniAPI API key instead.

Each successful endpoint-tool call uses the same credits as the matching direct REST API call and returns MCP metadata when available.

Recommended connection modes

Best option for most clients. Use the hosted MCP URL directly and sign in through the browser once. For clients that support remote MCP but not OAuth yet. Keep the server remote and send your KonbiniAPI API key as a Bearer token header.

Setup guides

Choose the client or tool you want to use with KonbiniAPI's hosted MCP server. These guides are grouped by use case and focus on the easiest remote setup with https://mcp.konbiniapi.com.

Chat

Set up with ChatGPT. Set up with Claude Desktop or Claude.ai. Connect your own agent or framework.

Automation

If you use KonbiniAPI inside workflow builders or agent orchestration tools, use the hosted MCP directly there too.

Connect KonbiniAPI to Zapier with MCP Client. Use KonbiniAPI inside n8n workflows or AI agents. Call KonbiniAPI tools from Make scenarios and agents. Add KonbiniAPI to Pipedream workflows and app integrations.

Code

Set up with Claude Code. Set up with OpenAI Codex CLI or the Codex app. Set up with Cursor. Set up with VS Code agent mode or GitHub Copilot. Set up with Antigravity. Set up with the Gemini CLI. Set up with Windsurf. Set up with the Cline VS Code extension. Set up with Replit Agent and Replit's MCP settings. Set up with Zed's agent panel and custom servers.