LurkAPI reviews
LurkAPI fits readers who need public social media data in a form that code or an AI app can consume without a lot of glue around each platform. Its core offer is broad coverage: it returns posts, ads and profiles from 16 platforms as clean JSON, and the docs show endpoints for Facebook, Reddit, YouTube, TikTok, Google, Bluesky, Snapchat, Truth Social, Telegram, X, Threads, Instagram, LinkedIn, link-in-bio pages, eBay and web pages. The product is built around simple GET requests, and the same API can be used through MCP in apps such as Claude, ChatGPT, Cursor and VS Code.
For readers trying to decide whether it matches a specific research job, the useful question is less “does it do social data?” and more “which kinds of social data does it cover well enough for my use case?” The documented endpoints include Facebook Ad Library search by keyword, with copy, media and landing page details; Google News search results with headline, publisher and publish time; Google Maps places with address, phone, rating, hours and coordinates; LinkedIn ads and jobs; TikTok videos, ads and hashtag metrics; Reddit posts; active eBay listings; and public posts from Instagram, X, Threads and Truth Social with engagement counts. That makes it a reasonable fit when the task is to fetch live public content into a downstream workflow, compare what is being advertised, or feed structured social data into an agent that can choose endpoints on demand.
The buying questions mostly come down to cost and usage shape. LurkAPI uses credits rather than a subscription, with new accounts getting 100 free credits and daily top-ups available when the balance falls under 25. The pricing page shows packs from $9 for 10,000 credits, and larger packs cost less per 1,000 credits. Most endpoints cost 1 credit, while some endpoints and extra steps cost more; the docs also show that web-page and speech-to-text endpoints can cost more depending on what is required. If a reader expects occasional API calls or wants to prototype before committing, the free credits and pay-per-call model are straightforward. If the work involves sustained volume, the credit model matters more than the headline platform count, because the right estimate depends on which endpoints are used.
One part of the product that may matter to AI-heavy teams is the MCP support. The docs describe a setup where an agent can ask for a platform-specific task in plain English and then call the right endpoint, and they also provide a compact mode for clients that load many tools up front. That is useful when the goal is not just to fetch data manually, but to let an AI app select among social-data endpoints and return JSON without building custom routing for each platform.
The main limit to keep in mind is scope: this is a public-data API. The official materials focus on public posts, ads, profiles, listings and search results, along with documented fields for those endpoints. If a project needs private data, broad account management, or a feature that is not listed in the docs, that is not part of the documented case for this product. For readers whose buying decision depends on one platform or one specific field, the endpoint pages matter more than the homepage pitch, because the practical fit is determined by the exact data each endpoint returns.