All templates
FeaturedAI#Drive#PDF#Agent#Skills#Document Q&A#Data#AI

Chat with Drive Docs

Ask questions about your Google Drive PDF files. An agent lists your Drive, reads the file you need, extracts its text, and answers in plain language.

Workflow at a glance

The full canvas, before you import it

Click any node to see its config.

#Drive#PDF#Agent#Skills#Document Q&A#Data#AI

Click a node to select it — same as the Heym editor; the panel shows its settings.

5 nodes · Free & source-available

Chat with Drive Docs

Connect your Google Drive and ask any question about your PDF documents. The agent automatically lists available files, reads the one you ask about, and extracts the text before answering.

How it works

  1. User query — type your question about a document.
  2. listDriveFiles tool — the agent calls the Drive node to see all available files.
  3. readDriveFile tool — the agent fetches the selected file as a base64-encoded binary.
  4. pdf_base64_to_text skill — a Python skill decodes the base64 string and extracts text from every page using pypdf.
  5. Answer — the agent analyzes the extracted text and replies.

Setup

  1. Open the docAgent node and pick an LLM credential.
  2. Make sure both Drive nodes are connected to the same Google Drive account via your credentials.
  3. Trigger the workflow from the Chat Portal or run it manually.

Notes

  • Supports multi-page PDFs; pages are joined with double newlines.
  • Scanned PDFs without a text layer will return an error — use an OCR step for those.
  • The Drive getAll node is capped at 1 result by default; raise the limit if you want the agent to browse more files.

How to import this template

  1. 1Click Import → Copy JSON on this page.
  2. 2Open your Heym and navigate to a workflow canvas.
  3. 3PressCmd+V/Ctrl+V— nodes appear instantly.
  4. 4Add your API keys in the node config panels and click Run.
More workflow templates
View all templates
Heym
incident analysis · production AI
Observed across 100s of AI rollouts

AI workflows don't fail because of prompts.
They fail because of orchestration.

symptom · glue code01
5 tools
Scripts, vector DB, approval bot, tracing, browser runner — none of them talk.
symptom · visibility02
~0%
Observable behavior across the stack. Debugging is guesswork.
with heym · one runtime
1 canvas
Agents, RAG, HITL, MCP, traces & evals. Self-hosted. Observable.
AI-Native RuntimeProduction-Grade
github.com/heymrun/heym