How-to guides
Each guide names the access constraint, the route that fits and the step where the workflow usually breaks.
How to extract YouTube transcripts in a workflow
The official captions API only covers videos you can edit. For public transcripts, use a schema-backed tool and keep the workflow honest…
Checked 2026-11-12How to give Claude access to LinkedIn data
The realistic routes for LinkedIn data in an agent: partner access, public company pages, and a hosted Actor with a published schema. No…
Checked 2026-11-12How to scrape a website with n8n
The n8n architecture that survives: a trigger, an input payload, a hosted Actor call, a normalisation step and a destination. Plus the…
Checked 2026-11-12n8n vs Make for scraping workflows
A practical comparison for data extraction: self-hosting, HTTP control, error handling, cost model and the limits that matter when a…
Checked 2026-11-12Reddit monitoring after the public API closes
The timeline, the official alternatives and the practical path for public community monitoring when RSS and the public Data API are gone.
Checked 2026-11-12How to schedule a scraper in n8n without duplicate rows
Set a schedule, fetch new rows, deduplicate with a stable key and handle the overlap between runs. The part most tutorials skip.
Checked 2026-11-12How to use Apify Actors with Claude Code
Install the MCP config, pick a schema-backed Actor, and prompt Claude Code for a structured dataset instead of a vague web summary.
Checked 2026-11-12Website to markdown for RAG: a clean ingestion route
How to turn public documentation into clean markdown for a retrieval corpus, with scope, deduplication and refresh rules that keep the…
Checked 2026-11-12