Personal
YouTube / RSS / PDF → scheduled Markdown knowledge base
Links, channels, and papers pile up unless something fetches them on a schedule and files Markdown.
What hurt
Mira (@MiramMint) built a knowledge pipeline with Grok Bot from Germany: it takes a YouTube link, channel, or RSS feed, pulls new items on a schedule, fetches research/PDF papers, processes them, and stores everything as Markdown files. Tip only — pipeline shape, no invented prompt. Not affiliated with xAI.
What to install or create
Name feeds. Schedule a pull. Fetch/process. Store Markdown. Read-only until you say a file may leave the vault.
Bots
Knowledge pipeline
Pulls YouTube / RSS / PDFs on a schedule. Writes Markdown. Does not publish.
Plugins
YouTube
Optional
Web search
Optional
Files
Markdown vault: one file (or folder) per pulled item.
Routines
Scheduled pull of the named YouTube / RSS / PDF sources into Markdown.
The prompt
This is a tip, not a pasteable system prompt. The author did not publish one. Follow the first-run steps.
First run
- 1Read https://x.com/MiramMint/status/2100692638353576192. Tip only — no invented summarizer prompt.
- 2Give the bot one YouTube link or channel, or one RSS feed, plus any PDF/research URLs you already trust.
- 3Ask it to fetch and process those items into Markdown files you can open.
- 4If the first run looks right, hang a schedule for new items. Prefer a coarse cadence.
- 5Keep the vault read-only to the outside. Publish or email still needs your yes.
Approval fence
Fetch and file Markdown only. No publish, email, or spend. Review new files before they leave the vault.
What to tweak
Steal feed → schedule → Markdown vault. Do not invent a summarizer prompt she did not post.
Source
Mira @MiramMint · Original post
Not affiliated with xAI. We did not write this prompt. Third-party template terms still apply.
Pro pack
Want the full setup, not just the idea?
Pro packs give you the job text, prompts, schedules and pitfalls for popular use-cases.