Templates / Screen CVs for hidden prompt injections with Gemini, Google Drive, Sheets, and Telegram

Screen CVs for hidden prompt injections with Gemini, Google Drive, Sheets, and Telegram

Ready for 3.0

by iamvaar ✓ · 2 views · created 2026-08-28 · HR, AI Summarization

Ready for 3.0. This template uses no nodes that n8n 3.0 removes. No removed nodes, versions or helpers found. Static check, last run 2026-10-19.

What this workflow does

This AI workflow has 20 nodes and runs when someone submits an n8n form (“Job Application”). Along the way it connects Google Gemini Chat Model, Google Drive, Telegram and Google Sheets. It also calls v2.convertapi.com over HTTP.

How it runs

Trigger: When someone submits an n8n form (“Job Application”).

  1. Check CV Format (If): checks a condition and branches into true / false
  2. Extract Text from PDF (Extract from File): extracts data from a file (pdf)
  3. Convert PDF to Image (HTTP Request): calls v2.convertapi.com (POST)
  4. Combine Screeners Results (Merge): merges the data from its input branches
  5. Notify Incorrect Format (n8n Form): shows the next form page
  6. Text Analysis Screener (Basic LLM Chain): sends a prompt to an LLM, with model: Gemini Text Chat; output parser: Parse Text Content
  7. Decode Image from Base (Convert to File): converts data to a file (to binary)
  8. Alert Vision Path Failure (Telegram): uses Telegram
  9. Set Verdict Fields (Edit Fields (Set)): sets 8 fields: text_score, vision_score, text_qualified…
  10. Vision Analysis Screener (Basic LLM Chain): sends a prompt to an LLM, with model: Gemini Vision Chat; output parser: Parse Visual Content

…plus 5 more steps. See the full node list below.

Services it uses

Google Gemini Chat ModelGoogle DriveTelegramGoogle Sheets

HTTP calls to: v2.convertapi.com

Before you upgrade to n8n 3.0

Nothing in this workflow is removed in n8n 3.0, so it should import and run after the upgrade.

Nodes used (14)

From the author

Quick overview Youtube Video: https://youtu.be/WMB6WwJayc This workflow collects PDF CVs via an n8n form, screens them twice with Gemini (text extraction and vision on a rendered image), and compares the results to detect hidden-text, prompt-injection…

Excerpt from iamvaar's description on n8n.io.

View the original on n8n.io → Check your own copy

Get the weekly “what broke” email

New templates, newly broken ones, and n8n release changes. One short email a week.

Form not loading? Subscribe here.