Agents

n8n

Read social data in an n8n workflow with the HTTP Request node, or give an n8n agent the MCP server.

Option A: HTTP Request node

  1. Add an HTTP Request node.
  2. Method GET, URL https://api.openhandle.dev/v1/instagram/profiles/@handle.
  3. Authentication: Generic Credential Type, Header Auth. Name Authorization, value Bearer followed by your key.
  4. Run it. The node output is the Openhandle envelope, with data holding the profile.

Pass freshness=24h as a query parameter for daily jobs. A cache hit costs $0.0005 and is often fresh enough for tracking.

Option B: MCP Client node

n8n's AI Agent can use MCP servers through the MCP Client tool node.

  1. Add an AI Agent node and attach an MCP Client tool.
  2. Endpoint https://api.openhandle.dev/mcp, transport Streamable HTTP.
  3. Header Authorization: Bearer followed by your key.
  4. Prompt the agent in plain words, for example "get the last five posts of this TikTok creator and sum the views".

Notes

  • Start with a Test key. Every endpoint works, nothing is charged.
  • Pagination is cursor based. The response's meta.cursors.next goes back in as cursor for the next page.
  • Errors come back with a code and a message. Failed requests are not billed.

LangChain

Load Openhandle tools into a LangChain agent through the MCP adapters, or call the REST API from a plain tool.

On this page

Option A: HTTP Request nodeOption B: MCP Client nodeNotes