Skip to content
Integrations

Langflow

Add Scrape.ly tools to Langflow flows through the MCP server.

Add Scrape.ly tools to a Langflow flow: an Agent component can search the Store, run Actors, wait for their runs, and read the results. Langflow connects to the hosted MCP server, so there is nothing to install beyond Langflow.

Create an API token in the dashboard under Settings → API & Integrations and set it as SCRAPELY_TOKEN. The flow can do anything the token can, including starting paid runs, so give it a scoped key: actors:run, runs:read, and datasets:read, with a Max charge per day. See API keys & permissions.

Add the MCP server

  1. In Langflow, open Settings → Global Variables and add a Credential variable named SCRAPELY_TOKEN. Set its value to Bearer followed by your token.
  2. Open Settings → MCP Servers and click Add MCP Server.
  3. Choose HTTP/SSE, name it scrapely, and set the URL to https://mcp.scrape.ly.
  4. Under Headers, add the key Authorization with the value SCRAPELY_TOKEN (the variable's name). Click Add Server. Langflow lists the server's tools when it connects.

Use the tools in a flow

  1. Add an MCP Tools component and set MCP Server to scrapely. Leave Tool empty to give the agent every tool.
  2. Turn on Tool Mode on the component and connect its Toolset output to the Tools input of an Agent component, with Chat Input and Chat Output around the agent.
  3. Open the Playground and ask:
.text
Find an Actor that scrapes Google Maps, run it for 'coffee shops in Berlin' with at most 20 results, and list the name, rating, and address of each place.

See Use Langflow as an MCP client for more options.

What the agent can do

The agent gets the MCP server's default tools. The ones used most:

ToolWhat it does
search-actorsFind Actors in the Store by keywords.
fetch-actor-detailsRead an Actor's input schema, README, and pricing before running it.
call-actorRun an Actor with a JSON input, optionally with maxItems and maxTotalChargeUsd.
get-actor-runWait for a run and check its status.
get-actor-outputRead a run's results, paged and kept compact.

The full list, including saved inputs and automations, is on the MCP server page. To run one Actor you already know, name it in the prompt, for example Run username/actor-name, so the agent skips the search.

Costs: runs started by the agent are billed like any run. Put the limits in the prompt (such as "at most 20 results"), and keep a Max charge per day on the key.