ScrapeIO

ScrapeIO returns public social media data as the same JSON shape on every platform. Every call spends credits; cached and failed calls are free.

Setup

The key is in the SCRAPEIO_API_KEY environment variable. If it is missing, ask the user to create a free key at https://scrapeio.com/login and set it. Never print the key.

Two ways to call

1. MCP tools (preferred when the scrapeio MCP server is connected): call the platform tool directly, for example tiktok_profile with { "handle": "nike" }. For anything without its own tool, use list_endpoints to find a path, then call_endpoint.
2. HTTP: fetch https://scrapeio.com/llms.txt for the endpoint list (or https://scrapeio.com/llms-full.txt for every parameter), then:

curl -s "https://scrapeio.com/v1/tiktok/profile?handle=nike" -H "x-api-key: $SCRAPEIO_API_KEY"

Rules

• Choose the narrowest endpoint for the question. Prefer handle when the user gives a username.
• The answer is in data. Lists have data.items, data.cursor and data.has_more.
• Fetch more pages only when the user needs them; each page is a separate paid call.
• credits_used and credits_remaining come back on every response. Tell the user when credits run low.
• 403 means private and 404 means not found: report it, do not retry. 502 means every source is down: retry once after a few seconds. 402 means the account needs credits (https://scrapeio.com/pricing).

Report

Lead with the numbers the user asked for. Name posts by caption or title and date, and link to them with the url field. Field names match across platforms, so one table can compare TikTok, Instagram and YouTube side by side.