Organize new PDFs with n8n and save the findings
An n8n flow watches a Drive folder, pulls details from each new PDF, and records the paper in Sheets and Notion.
- What they did
- They wired Google Drive to start the workflow whenever a PDF was added to a specific folder. After downloading the file, a PDF Vector community node extracted the paper’s metadata, abstract, study type, and findings. The workflow then used the paper’s 3 main keywords to look up related work in Semantic Scholar and PubMed, generated an APA-style citation, wrote a line to Google Sheets, and made a Notion page.
- How it went
- A single paper usually finished in 20-30 seconds. The setup reached about 98% accuracy for metadata on digital PDFs, around 90% for study type, and about 85% on scanned files. On narrow subjects, the search could surface only 3-5 related papers, and the citation still needed review before use.
- Tips
- This only works with self-hosted n8n, and PDF Vector uses roughly 4-5 credits per paper. That means the free 100 credits are enough for about 20-25 papers each month, so budget for usage and verify the APA citation before relying on it.
Try it yourself with n8n
Set up an n8n workflow that starts when a new PDF appears in [Google Drive folder]. For each file, download it, pull out the title, authors, year, abstract, study type, and key findings, then search Semantic Scholar and PubMed with the paper’s top 3 keywords. Draft an APA citation, add the details to [Google Sheets name], and create a [Notion page/database] entry with a short plain-English summary and the related papers. Before any action that would send, publish, submit, delete, or pay for anything, stop and ask me to approve the exact step. After the first test PDF, show me the sheet row and the Notion page so I can confirm it worked.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Save cleaned web pages as structured Notion notes
They set up n8n to grab a page from a URL, extract readable text, summarize it with OpenAI, and store the output in Notion.
Automate daily arXiv paper tracking in Notion
They used n8n to pull fresh arXiv results, condense each paper with Gemini, and file the entries in Notion with email and Feishu alerts.
Build a Notion research report from a topic
An n8n flow asks a few setup questions, gathers source text, writes a report, and posts it to Notion.
Turn Telegram links into a searchable second brain
An n8n Telegram bot saves links to Sheets, keeps one record per URL, and answers questions by searching stored summaries and page text.
Summarize Reddit discussions into Notion pages
An n8n flow collects Reddit threads, lets Gemini pull out themes, and posts the findings into Notion.
Use Codex to summarize Logseq journals from the CLI
They used Codex with the Logseq CLI docs to query a note graph and turn months of journals and tagged pages into summaries.