# Dify (/docs/integrations/dify) Pull social media data into Dify without writing a server. You can call SocialCrawl from a Dify workflow with an HTTP Request node. One GET request returns structured data for the next node. Setup takes three steps: set the header, add the node, route the failure. ## Before you start Create an API key in the dashboard under API Keys. The HTTP Request node has its own authorization setting, so enter the key there and not in the URL or a prompt. Prove the key with a call that costs nothing: ```bash title="Check the key (0 credits)" curl "https://www.socialcrawl.dev/v1/credits/balance" \ -H "x-api-key: sc_your_key_here" ``` A reply with `"success": true` and your balance in `data.balance` means the key works. ## Connect Add an HTTP Request node. Set the method to `GET` and the URL to `https://www.socialcrawl.dev/v1/tiktok/profile?handle=charlidamelio`. For authorization, choose API Key with a custom header, name it `x-api-key`, and enter your key. To turn a whole endpoint into a tool the model can pick, Dify can also import an OpenAPI specification. Our specifications are listed in [AI agents](/docs/ai-agents.md). ## Read the answer Every response has the same shape. Branch on `success`. A list puts its rows in `data.items`, and `pagination.next_cursor` is the token for the next page. ```json title="Response (trimmed)" { "success": true, "data": { "author": { "username": "charlidamelio" } }, "credits_used": 1, "credits_remaining": 99 } ``` To read the next page of a list, call again with `next_cursor` as the `cursor` query parameter, and stop when `pagination.has_more` is `false`. See [Pagination](/docs/pagination.md). ## When it stops Dify nodes have an error handling setting with a fail branch. Turn it on for the HTTP Request node and connect the fail branch to an end node that returns the message, so a refused call reaches the user as text instead of a stopped run. When credits run out, the answer is a 402 with `error.type` set to `INSUFFICIENT_CREDITS`. Read `error.message` for the text and `error.details.checkout_url` for the link that opens checkout. When the link is absent, `error.details.top_up_url` opens the billing page. Never retry a 402, because it fails until credits are added. The node's retry setting suits a 429 or a 502, 503 or 504, and not a 402. For a 429, wait the seconds in the `Retry-After` header. See [Errors](/docs/errors.md). ## What a run costs It depends on the endpoint, and some endpoints are priced by what they return. [Endpoint pricing](/docs/endpoint-pricing.md) lists each one, and [Credits](/docs/credits.md) explains cache hits and refunds. A workflow that runs per user message calls the API each time. `GET /v1/utility/estimate` quotes a call for 0 credits. - [Errors](/docs/errors.md): Every error code, with the fix and whether to retry. - [AI agents](/docs/ai-agents.md): OpenAPI files, llms.txt and the other routes an agent can read. - [Endpoint pricing](/docs/endpoint-pricing.md): What each endpoint costs before a workflow runs it.